<%BANNER%>

Multiple Modeling and Control of Nonlinear Systems with Self-Organizing Maps


PAGE 1

MULTIPLE MODELING AND CONT ROL OF NONLINEAR SYSTEMS WITH SELF-ORGANIZING MAPS By JEONGHO CHO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by JEONGHO CHO

PAGE 3

This document is dedicated to my late father My inspiration and the one who gave me every opportunity to realize my dreams.

PAGE 4

ACKNOWLEDGMENTS Being at CNEL has been not only a wonderful academic experience but also a unique opportunity to meet many colleagues who helped me immensely along the way. It is a privilege to me to acknowledge the unconditional support of my supervisor, Dr. Jose C. Principe, who has been a mentor during my years as a graduate student. His advice, wisdom, and many invaluable lessons in life have made this dissertation possible. I sincerely appreciate the help offered by the members of my academic committee, Dr. John G. Harris, Dr. Michael C. Nechyba, and Dr. Loc Vu-Quoc. Their cooperation and suggestions have considerable improved the quality of my dissertation. I would also like to express my gratitude to Dr. Mark A. Motter and Dr. Deniz Erdogmus for their help and providing many valuable comments during the past years of my stay in the CNEL. My deepest recognition goes to my beloved parents, especially to my late father who helped me in any imaginable way to achieve my objectives and fulfill my dreams. They have been an inexhaustible source of love and inspiration all my life. My most special thanks go to my wife, Joonhee, for her patience, understanding and encouragement, without which it would have been impossible to complete this dissertation. Finally, I thank my specially beloved son, Minsuh, hoping that the effort of these years may offer him a more plentiful life in the years to come. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ............................................................................................................vii LIST OF FIGURES .........................................................................................................viii ABSTRACT .......................................................................................................................xi CHAPTER 1 INTRODUCTION........................................................................................................1 1.1 Motivations.............................................................................................................1 1.2 Review of Literature...............................................................................................2 1.3 Objectives and Authors Contribution....................................................................5 1.4 Outline....................................................................................................................6 2 SYSTEM IDENTIFICATION VIA MULTIPLE MODELS.......................................8 2.1 Local Dynamic Modeling.......................................................................................8 2.2 SOM-Based Local Modeling................................................................................10 2.2.1 Reconstruction of State-Space....................................................................11 2.2.1.1 Delay reconstruction........................................................................11 2.2.1.2 Estimating an embedding dimension...............................................12 2.2.2 The Self-Organizing Map...........................................................................14 2.2.2.1 Competitive process.........................................................................15 2.2.2.2 Cooperative process.........................................................................16 2.2.2.3 Adaptive process..............................................................................17 2.2.3 Modeling Methodology Based on the SOM...............................................17 2.3 Input-Output Representation of Systems..............................................................19 2.3.1 Classical Approach.....................................................................................20 2.3.2 Series-Parallel and Parallel Models............................................................21 2.3.3 SOM-based Multiple ARX Models............................................................23 2.3.3.1 Selection of operating regions with a SOM.....................................23 2.3.3.2 Model development procedure.........................................................27 3 MULTIPLE MODEL BASED CONTROL...............................................................29 v

PAGE 6

3.1 Discrete-Time Control System.............................................................................31 3.2 Inverse Control via Backpropagation Through Model.........................................32 3.3 Multiple Inverse Control.......................................................................................34 3.4 Multiple PID Control............................................................................................37 4 MULTIPLE QUASI-SLIDING MODE CONTROL.................................................42 4.1 Introduction to Variable Structure Systems..........................................................42 4.1.1 Sliding Hyperplane Design.........................................................................44 4.1.2 Sliding Mode Control Law Design.............................................................45 4.2 Sliding Mode Control in Sampled-Data Systems.................................................48 4.2.1 Quasi-Sliding Mode....................................................................................49 4.2.2 Quasi-Sliding Mode Control Using Multiple Models................................52 4.3 Analysis of Multiple Quasi-Sliding Mode Control with an Imperfect Sensor.....54 5 CASE STUDIES.........................................................................................................57 5.1 Controlled Chaotic Systems.................................................................................57 5.1.1 The Lorenz System.....................................................................................58 5.1.2 The Duffing Oscillator...............................................................................66 5.2 Nonlinear Discrete-Time Systems........................................................................72 5.2.1 A First-order Plant......................................................................................72 5.2.2 A Laboratory-scale Liquid-level Plant.......................................................78 5.3 Flight Vehicles......................................................................................................86 5.3.1 Missile Dynamics.......................................................................................86 5.3.2 LoFLYTE UAV..........................................................................................92 6 CONCLUSIONS AND FUTURE WORK...............................................................101 6.1 Summary.............................................................................................................101 6.2 Future Work........................................................................................................103 6.3 Concluding Remarks..........................................................................................104 APPENDIX QSMC FOR MIMO SYSTEM........................................................................................106 LIST OF REFERENCES.................................................................................................109 BIOGRAPHICAL SKETCH...........................................................................................118 vi

PAGE 7

vii LIST OF TABLES Table page 5-1 Lipschitz index of the controlled Lo renz system for determining an embedding dimension.................................................................................................................59 5-2 Comparison of modeling performance for the controlled Lorenz system...............62 5-3 Comparison of modeling performance fo r the controlled Duffing oscillator..........67 5-4 Comparison of tracking performance for 3 different control task : Settling time and NRMS-SSE .........................................................................................................70 5-5 Lipschitz index of a laboratory-scale liquid-level plant for determining an embedding dimension..............................................................................................79 5-6 Comparison of modeling performa nce for the liquid-level plant.............................80 5-7 Comparison of control performance fo r the liquid-level plant in noise-free environment..............................................................................................................83 5-8 Comparison of control performance for the liquid-level plant in the presence of sensor noise: standard devi ation of noise is 4.5e-2..................................................83 5-9 Comparison of modeling perfor mance for the missile system.................................87 5-10 Comparison of controller performance fo r the missile system in the presence of noise having the standard deviation of 4.6e-2..........................................................90 5-11 Comparison of modeling perfor mance for the lateral motion ( p and r ) of the LoFLYTE UAV.......................................................................................................95

PAGE 8

viii LIST OF FIGURES Figure page 2-1 Local modeling scheme on the basis of a SOM.......................................................10 2-2 Two data points that are close in 1 but distant in 2 .............................................13 2-3 Kohonens Self-organizing Map..............................................................................15 2-4 Nonlinear dynamic model configuration..................................................................20 2-5 A series-parallel model (left) and A parallel model (right)......................................22 2-6 Configuration of local li near modeling based on a SOM.........................................27 3-1 Classical discrete-time control system.....................................................................31 3-2 Modeling and control scheme using th e TDNN: (a) TDNN modeling of a plant (b) An inverse controller via Bac kpropagation through (Plant) Model...................34 3-3 Proposed SOM-based inverse control scheme.........................................................36 3-4 PID controlled system..............................................................................................38 3-5 Overall schematic diagram of the non linear PID closed loop control mechanism using multiple controllers.........................................................................................39 3-6 Block diagram of PID contro ller for a SISO plant model........................................40 4-1 Phase plane plot of a continuous-time second-order variable structure system.......46 4-2 Discrete-time system response with sliding mode control.......................................49 5-1 The uncontrolled Lorenz system: phase -space trajectory and time-series...............59 5-2 Generalization error v.s. Number of PEs (left) Learni ng curve (right)....................60 5-3 Identification of the controlled Lorenz system by multiple models.........................61 5-4 Tracking a fixed point re ference signal by MIC: (a) 0 dy (b) 8 dy ................63

PAGE 9

ix 5-5 Comparison of control performance vary ing the number of inverse controllers based on multiple models.........................................................................................64 5-6 Comparison of tracking performance by multiple model based controllers (MIC, MPIDC, MQSMC) and global invers e controllers (IC-ARX, TDNNC).................65 5-7 The uncontrolled Duffing oscillator: pha se-space trajectory and time-series..........66 5-8 Lipschitz index (left) for the determin ation of optimal number of inputs and outputs and Generalization error v.s. Number of PEs (right)...................................67 5-9 Identification of the controlled Duffi ng oscillator by TDNN (left) and multiplemodels (right)...........................................................................................................68 5-10 Performance comparison on trajector y tracking by TDNNC, PID-ARX, and MPIDC when the poles of the closed-loop response are place at (a) 0.9 (b) 0.5 0.5i (c) 0.25 0.25i (d) 0.05 0.05i................................................................69 5-11 Control performance by TDNNC (left) and MPIDC (right)....................................71 5-12 Parameter selection to design multiple models........................................................72 5-13 Modeling performance using 64 multiple models for a nonlinear first order plant................................................................................................................73 5-14 Responses for parameter selec tion to design QSMC by varying (a) r T and (b) qT. .................................................................................................74 5-15 Comparison of tracking performance usi ng a global controller and multiple model based controllers in the absence of se nsor noise. The figure (right) is an enlargement of the figure (left) between 34 and 52 iterations.................................75 5-16 Performance of square-wave tracking in the absence of noise by the MQSMC......76 5-17 Comparison of performance agains t noise among TDNNC, MIC, MPIDC and MQSMC. .................................................................................................................77 5-18 Sinusoidal and arbitrar y signal tracking by the MQSM C:(a) in the absence of sensor noise (b) in the presence of sensor noise, dB SNR 20 ...............................78 5-19 Modeling performance using multiple models for a liquid-level plant....................79 5-20 Typical input-output char acteristic of the second-order liquid-level plant..............81 5-21 Square-wave tracking performance of th e liquid-level plant varying the sliding surface and the noise level by the MQSMC.............................................................81

PAGE 10

x 5-22 Control of the liquid-level plant by the MQSMC varying the number of controllers: (a) M = 1 (b) M = 16 (c) M = 36 (d) M = 144.......................................82 5-23 Control of the liquid level system with measurement noise by the MQSMC with (a) M = 1 (b) M = 16 (c) M = 144 and (d) the TDNNC...................................84 5-24 Tracking an oscillatory reference signal of the liquid-level plant by the TDNNC (left) and the MQSMC (right) in the presence of sensor noise................................85 5-25 Performance assessment on a trajecto ry tracking under noisy environment...........85 5-26 Modeling performance using multiple models for the missile dynamics................87 5-27 Tracking various set-point reference signal by (a) TDNNC (b) MIC (c) MPIDC and (d) MQSMC in the absence of measurement noise...........................................88 5-28 Tracking various set-point reference signal by (a) TDNNC (b) MIC (c) MPIDC and (d) MQSMC under the presence of measurement noise whose standard deviation is 4.6e-2....................................................................................................89 5-29 Trajectory tracking by TDNNC (left) and MQSMC (right) in the presence of noise whose standard deviation is 2.3e-1.................................................................90 5-30 Set-point tracking beha vior by the TDNNC (left) and the MQSMC (right) under parameter variations.................................................................................................91 5-31 General description of ai rcraft (left) and LoFLYTE testbed UAV (right).............92 5-32 Control inputs (r a ) used to generate data samp les for training the networks....94 5-33 Modeling of a roll-rate using multiple models.........................................................95 5-34 Comparison for controlling roll-rate and yaw-rate to track the set point in the absence of noise by (a) TDNNC (b) MIC (c) MQSMC...........................................98 5-35 Comparison for controlling roll-rate and yaw-rate to track the set point in the presence of noise by (a) TDNNC (b) MIC (c) MQSMC.........................................99 5-36 Performance of controlling roll-rate and yaw-rate to track an arbitrary trajectory with measurement noise (SNR = 20) by (a) TDNNC (b) MIC (c) MQSMC.........100

PAGE 11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MULTIPLE MODELING AND CONTROL OF NONLINEAR SYSTEMS WITH SELF-ORGANIZING MAPS By Jeongho Cho December 2004 Chair: Jose C. Principe Major Department: Electrical and Computer Engineering This dissertation is concerned with the development and analysis of a nonlinear approach to modeling and control of nonlinear complex systems. In particular, the problem of designing a mathematical model of a nonlinear plant using only observed data is considered. For the identification of the plants, the concept of multiple models with switching is employed in order to simplify both the modeling and the controller design since a single controller may sometimes have difficulty meeting the design specifications in case the dynamics vary considerably over the operating region. For this reason, a Self-Organizing Map (SOM) is utilized to divide the operating region into local regions as a modeling infrastructure to construct local models. The SOM selects the local operating region relying on the embedded output, and the local model is built by the embedded output as well as the embedded control input data samples which are spaced in the local area. xi

PAGE 12

Based on the identified multiple models, the problem of designing controllers is discussed. Each local linear model is associated with a linear controller, which is easy to design. Switching of the controllers is done synchronously with the active local linear model that tracks different operating conditions. The effectiveness of the proposed approach is shown through experiments for modeling complex nonlinear plants such as chaotic systems, nonlinear discrete time systems and flight vehicles. Its comparison with neural networks-based alternatives, Time Delay Neural Network (TDNN), shows clear advantages of local modeling and control in terms of performance. xii

PAGE 13

CHAPTER 1 INTRODUCTION 1.1 Motivations The identification of nonlinear dynamical systems has received considerable attention since it is an indispensable step towards analysis, simulation, prediction, monitoring, diagnosis, and controller design for nonlinear systems [21,65,67]. In particular, the problem of designing a mathematical model of a nonlinear plant using only observed data has attracted much interest, both from an academic and an industrial point of view. During the past few years, neural networks as a global model have been suggested for nonlinear dynamical black-box modeling and successfully applied to the prediction and modeling of nonlinear processes [11,46,65,73]. Global models, however, have shown some difficulties in cases when the dynamical system characteristics vary considerably over the operating regime, effectively bringing the issue of time varying parameters (or nonlinearity) into the design. On the other hand, local modeling derives a model based on neighboring samples in the operating space to characterize some operating point or similar feature [23,85]. If a function f to be modeled is complicated, there is no guarantee that any given global representation will approximate f equally across all space. Moreover, nonlinear models are too complex to be used for controller design [70]. Thus, nonlinear control methods cannot serve all needs of real industrial control problems. In this case, the dependence on representation can be reduced using local approximation where the domain of f is divided into local regions and a separate model is 1

PAGE 14

2 used for each region [3,13,24]. In a number of local modeling applications, a Self-Organizing Map (SOM) has been utilized to divide the operating regions into local regions [30,60,74,97]. The SOM is particularly appropriate for switching, because it converts complex, nonlinear statistical relationships of high-dimensional data into simple geometric relationships that preserve the topology in the feature space [44]. Thus the role of the SOM is to discover patterns in the high dimensional state space and divide that space into a set of regions represented by the weights of each Processing Element (PE). Linear models and associated techniques for linear control design are typically used to control the plant under certain specific operating conditions. This type of control is only valid in a small region around the operating point. For that reason, the concept of multiple models with switching, according to a change in dynamics, has been an area of interest in control theory in order to simplify both modeling and controller design [62,66]. The motivation for this research, therefore, is to explore control strategy using SOM-based multiple models for nonautonomous and nonlinear systems. 1.2 Review of Literature There are many examples in the literature in which the local modeling paradigm has been successfully applied for the modeling of nonlinear autonomous and nonautonomous systems. Farmer and Sidorowich [23] have shown that local linear models, despite their simplicity, provide an effective and accurate approximation of chaotic dynamical systems. Jacobs et al. [38] have proposed the mixtures of expert models that are composed of several different expert networks and a gating network that localizes the experts. They showed that a simple model can be built by dividing a vowel discrimination task into appropriate subtasks. Bottou and Vapnik [3] have proposed using local learning algorithms instead of training a complex system with all data samples, and

PAGE 15

3 demonstrated that a set of subsystems trained with a subset of data can improve the performance for an optical character recognition problem. The neural-gas architecture proposed by Martinetz et al. [53] is similar to the SOM in that the competitive network divides the input space in a set of smaller regions and then local linear models are created by a LMS-like rule. They showed that the neural-gas network outperforms MLPs and RBF networks for time series prediction. The same group [77] used a SOM for the control of a robotic arm. Murray-Smith et al. [61] similarly have extended RBF networks where each local model is a linear function of the input and exhibited great success in control problems. Principe and Wang [74] have successfully modeled a chaotic system with a SOM-based local linear modeling method. Vesanto [94] and Moshou and Ramon [60] proposed a scheme that essentially followed local linear modeling based on SOM topology for nonautonomous system. Under some conditions, it has been shown that multiple models can uniformly approximate any system on a closed subset of the state space provided a sufficient number of local models are given. Generally, the control using multiple models is categorized by two approaches: a global model-based control using local models and a multiple model-based control with switching. Global controller design with the aid of multiple linear models has been extensively reported in the literature. Gain scheduling has been perhaps the most common systematic approach to control nonlinear systems in practice due to its simple design and tuning [47,68,79]. The multiple model adaptive control approach differs from gain scheduling mainly in the use of an estimator-based scheduling algorithm used to weight the local controllers. Murray-Smith and Hunt [61] utilized an extended RBF network where each local model is a linear function of the input and reported great

PAGE 16

4 success for control problems. The overall controller designed is based on the local models and a validity function to guarantee smooth interpolation. Similarly, Foss et al. [24] and Gawthrop and Ronco [29] employed model predictive controllers and self-tuning predictive controllers, respectively, using multiple models. Palizban et al. [70] attempted to control nonlinear systems with the linear quadratic optimal control technique using multiple linear models and provided the stability condition for the closed loop system. Ishigame et al. [37] proposed the sliding mode control scheme based on fuzzy modeling composed of a weighted average of linear systems to stabilize an electric power system. In contrast, Narendra et al. [66] proposed the multiple model approach in the context of adaptive control with switching where local model performance indices have been used to select the local controller. Subsequently, Narendra and Balakrishnan [64] proposed different switching and tuning schemes for adaptive control that combine fixed and adaptive models yielding a fast and accurate response. Principe et al. [75] proposed a SOM-based local linear modeling strategy and predictive multiple model switching controller to control a wind tunnel and showed improved performance with decreased control effort over both the existing controller and an expert human-in-the-loop control. Later, Narendra and Xiang [63] proved that the adaptive control using multiple models is globally stable and that the tracking error converges to zero in the deterministic case. Diao and Passino [20] applied multiple model based adaptive schemes to the fault tolerant engine control problem. A linear robust adaptive controller and multiple nonlinear neural network based adaptive controllers were exploited by Chen and Narendra [11]. Thampi et al. [89,90] have also shown the applicability of the multiple model approach based on the SOM for flight control.

PAGE 17

5 1.3 Objectives and Authors Contribution This dissertation is concerned with modeling and control of nonlinear nonautonomous dynamical systems. The objective of this dissertation is to investigate if it is possible to obtain a better result in extending the formulation of the control problem from using just one global model to using several internal models. Thus a multiple modeling approach is presented and techniques to design controllers based on these model structures are developed. The main contributions made by the author with respect to the modeling and control of nonlinear systems include: Firstly, an extended version of the SOM-based local modeling scheme for nonautonomous and nonlinear plants is developed for more general representation of the underlying dynamical systems and better approximation solely based on input-output measurements of the plant. Local linear models are derived through competition using the SOM and they are derived from the data samples corresponding to each of the SOMs PEs. Secondly, we investigate several options regarding how to capture the dynamics in the input-output joint space. It is shown that as the number of dependent variables is increased SOM modeling may become increasingly difficult to model accurately due to its memory based structure. Thus, the SOM is trained to position the local models in the embedded output space. At any time instant, the model representing the plant dynamics is chosen by the SOM depending on the history of the plant and then incorporated with the previous control inputs. Thirdly, the model structure is controller oriented since the dynamics are simpler locally than globally such that it is easier to develop local models as well as controllers.

PAGE 18

6 For instance, if the system phenomena or behavior changes smoothly with the operating point, then a linear model (or controller) will always be sufficiently accurate locally provided that the operating region is sufficiently small, even though the system may contain complex nonlinearities when viewed globally. These local controllers can then be switched as the system changes operating conditions. Hence, multiple control with switching, such as an inverse controller and a PID controller using identified multiple models, are examined. Finally, in order to obtain a controller which preserves the good sensitivity to external disturbances, a sliding mode controller (SMC) is employed using multiple models. By doing so, one of the difficulties in designing a SMC (that requires the complete knowledge of the plant to be controlled) can be removed. In addition, we examine the effect by the modeling error due to the quantization of state space as well as by measurement noise to the proposed multiple model based sliding mode control performance. It is shown that the switching scheme does not create an issue to be considered in order to guarantee BIBO stability of the overall system. 1.4 Outline This dissertation is divided into six chapters. Chapter 2 gives a review of the local dynamic modeling required to study further for nonautonomous systems. The SOM employed as a modeling infrastructure to construct the local models is described briefly. At the end, the proposed SOM-based multiple ARX modeling scheme is introduced for nonlinear nonautonomous dynamic systems representation. Chapter 3 shows how to design an inverse control and a PID control framework based on the designed multiple linear models. A description of a global nonlinear TDNN trained by backpropagtion through a model is also presented for comparisons regarding performance. Chapter 4

PAGE 19

7 gives a brief description of Variable Structure Systems (VSS). Quasi-sliding mode control strategy is proposed based on multiple models. In addition, analysis of multiple quasi-sliding mode control structure with an imperfect sensor is discussed. In Chapter 5, simulations are conducted assuming that both the plant is unknown and the only state available for measurements is the plant output for two controlled chaotic systems, two nonlinear discrete-time systems, one missile, and one Unmanned Aerial Vehicle (UAV). Finally, Chapter 6 presents conclusions based on the preceding analysis and simulation results and suggests further study.

PAGE 20

CHAPTER 2 SYSTEM IDENTIFICATION VIA MULTIPLE MODELS The idea of multiple modeling is to approximate a nonlinear system with a set of relatively simple local models valid in certain operating regimes [39]. Because of the complexity, uncertainty and nonlinearity of a large class of systems, we often cannot derive appropriate models from first principles, and are not capable of deriving accurate and complete equations for input-state-output representations of the systems. Hence we need to resort to input-output data in order to derive the unknown nonlinear system model [10,35]. The technique of multiple model networks is appealing for modeling complex nonlinear systems due to its intrinsic simplicity [62,66]. 2.1 Local Dynamic Modeling We begin with a brief overview of a dynamical systems approach to input-output modeling. When no physical knowledge of the system is available, we have to determine a model from a finite number of measurements of the systems inputs and outputs. An autonomous dynamical systems approach to black-box modeling based on Takens Embedding theorem was first suggested by Casdagli [7]. The delay embedding offers the possibility of accessing linear or nonlinear coupling between variables and is a fundamental tool in nonlinear system identification. The use of delay variables in the structure of these dynamical models is similar to that originally studied by Leontaritis and Billings [49], and is common in linear time-series analysis and system identification [96]. When we are trying to understand an irregular sequence of measurements, an immediate question is what kind of process generates such a series. Under the 8

PAGE 21

9 deterministic assumption, irregularity can be autonomously generated by the nonlinearity of the intrinsic dynamics. Let the possible state of a system be represented by points in a finite dimensional phase space, This can be realized by a map of onto itself: x P P 121PkkPkkxxfxx (2.1) The predictive mapping is the centerpiece of modeling since once determined, f can be obtained from the predictive mapping as Pif: )(,1kxikfx (2.2) where TPkkkkxxxx],,,[11, In addition, Singer et al. [85] derived the locally linear prediction based on this relationship as ikxTikxibaf,,)( (2.3) The vector and scalar quantities of a and are estimated from the selected pairs ( b jxjx,1, ) in the least square sense, where j is the index of the data samples in the operating regime, i.e., one model. To obtain a stable solution, more than P pairs must be selected. In general, the above local model fitting is composed of two steps: a set of nearby state searches over the signal history and model parameters which, when pieced together, provide a global modeling of the dynamics in state space. The underlying dynamics f is then approximated as (2.4) iNiff,,1 where N is the number of operating regimes. Based on this approximation of an autonomous system, local linear models have performed very well in comparative studies

PAGE 22

10 on time series prediction problems and in most cases have generated more accurate predictions than global methods [84,97]. Moreover, the nonlinear dynamical system can be identified by local framework even in the presence of noise if enough data are available to cover all of the state space since local regions are local averages of the data. To make the local network less sensitive to noise and outliers, more than one neighbor can be utilized in local modeling. 2.2 SOM-Based Local Modeling The SOM is employed as a modeling infrastructure to construct the local models. It provides a codebook representation of the plant dynamics and organizes the different dynamic regimes in topological neighborhoods. Thus we can create a set of models that are local to the data in the Voronoi tessellation created by the SOM. This local model structure with the SOM is depicted in Figure 2-1. ku Figure 2-1. Local modeling scheme on the basis of a SOM. 1ky Switching Device (SOM) Reconstruction of State-Space M ode l 1 M odel 2 Plant M odel N ky

PAGE 23

11 2.2.1 Reconstruction of State-Space In many cases of practical interest it is not possible to measure the state variables of a system directly. Instead, the measuring procedure yields some value )(kkxy when the system is in states Here, kx )( is a measurement function which in general depends on the state variables in a nonlinear way. The time evolution of the state of the system results in a scalar time series In order to reconstruct the underlying dynamics in phase space, delay embedding techniques are commonly used. ,,,321xxx 2.2.1.1 Delay reconstruction Delay-coordinate embedding [41,91], a technique developed by the dynamics community, is one way to help the input-output modeling; it allows one to reconstruct the internal dynamics of a complicated nonlinear system from a single time series. That is, one can often use delay-coordinate embedding to infer useful information about internal (and immeasurable) states using only output information. The reconstruction produced by delay-coordinate embedding is not, of course, completely equivalent to the internal dynamics in all situations, or embedding would amount to a general solution to control the theorys observer problem: how to identify all of the internal state variables of a system and assume their values from the signals can be observed [87]. However, a single-sensor reconstruction, if done properly, can still be extremely useful because its results are guaranteed to be topologically (i.e., qualitatively) identical to the internal dynamics. This means that conclusions drawn about the reconstructed dynamics are also true of the internal dynamics of the system inside the black box. In order to reconstruct the underlying dynamics in phase space, we begin with scalar observable, of the state ky kx

PAGE 24

12 of a deterministic dynamical system. Then typically we can reconstruct a copy of the original system by considering blocks TdykkkTdykkkkyyyyxxx )1()1(,,,,)(,),(),( (2.5) of successive observations of dy for sufficiently large. Delay-coordinate reconstruction is governed by two parameters, embedding dimension and embedding delay dy dy Note that using 1 dy merely returns the original time series; one-dimensional embedding is equivalent to not embedding at all. Proper choice of and dy is critical to this type of phase-space reconstruction and must therefore be done wisely; only correct values of these two parameters yield embeddings that are guaranteed by the Takens theorem [88] and subsequent work by Packard et al. [69] and Casdagli et al. [7] to be topologically equivalent to the original (unobserved) phase-space dynamics. 2.2.1.2 Estimating an embedding dimension There has been much work on determining the embedding dimensions of the time series generated by autonomous dynamical systems in the absence of dynamical noise [41,91,42,6]. The methods developed for estimating the minimum embedding dimensions are grounded on Takens embedding theorem [88] and most of them use the ideas of the false nearest neighbors technique [42,6]. Later a number of works discussed theoretical foundations of the delay embedding of the input-output time series [7,6]. This led to the generalization of the existing method for the case of non-autonomous dynamical systems [87,6,76]. He and Asada [35] proposed a strategy which is based directly on measurement data and does not make any assumptions about the intended model architecture or

PAGE 25

13 structure. It requires only that the process behavior can be described by a smooth function, which is an assumption that must be made in black box nonlinear system identification. An explanation of this strategys central idea follows. In general case, the task is to determine the relevant inputs of the function ),,,(211nkfy (2.6) from a set of potential inputs )(,,,21noo that is given. If the function in (2.6) is assumed to depend on only 1 n inputs although it actually depends on n inputs, the data set may contain two (or more) points that are very close (in the extreme case they can be identical) in the space spanned by the 1 n inputs but differ significantly in the nth input. This situation is shown in Figure 2-2 for the case 2 n 2 iy jy 1 Figure 2-2. Two data points that are close in 1 but distant in 2 The two points i and j are close in the input space spanned by 1 alone but they are distant in the 21 input space. Because these points are very close in the space spanned by the inputs ( 1n 1 ) it can be expected that the associated process outputs and are also close (assuming that the function iy jy )( f is smooth). If one (or several) relevant inputs are missing then obviously and are expected to take totally iy jy

PAGE 26

14 different values. In this case, it is possible to conclude that the inputs are not sufficient. Thus, the nth input should be included and the investigation may begin again. 1n In [35] an index is defined based on so-called Lipschitz quotients, which is large if one or several inputs are missing (the larger the quotients, the more inputs are missing) and is small otherwise. Thus, using this Lipschitz index the correct embedding dimensions can be detected at the point where the Lipschitz index ceases to decrease. The Lipschitz quotients in the one-dimensional case are defined as jijiforyylLjijiij,, (2.7) where is the number of samples in the data set. For the multidimensional case, the Lipschitz quotients can be calculated by the straightforward extension of (2.7): L jijiforyylLjninjijinij,,)()(2,,2,1,1 (2.8) where is the number of input. The Lipschitz index, then, can be defined as the maximum occurring Lipschitz quotient n )(max)(,,nijjijinll (2.9) As long as n is too small and thus not all relevant inputs are included, the Lipschitz index will be large because situations as shown in Figure 2-2 will occur. As soon as all relevant inputs are included, (2.9) stays relatively constant. 2.2.2 The Self-Organizing Map The principal goal of the SOM is to transform an incoming signal pattern of arbitrary dimension into a one or two-dimensional discrete map, and to perform this transformation adaptively in a topologically ordered fashion [44]. Figure 2-3 shows

PAGE 27

15 Kohonens model of a two-dimensional SOM. Each PE in the lattice is fully connected to all the source PEs in the input layer. This network represents a feedforward structure with a single computational layer consisting of PEs arranged in rows and columns. Figure 2-3. Kohonens Self-organizing Map. The algorithm responsible for the formation of the SOM proceeds first by initializing the synaptic weights in the network. Once the network has been properly initialized, there are three essential processes involved in the formation of the SOM: competition, cooperation and adaptation. Descriptions of these processes follow. 2.2.2.1 Competitive process Let denote the dimension of the input space. Let an input vector selected randomly from the input space be denoted by m Tkmkkkxxxx,,2,1,,, (2.10) The synaptic weight vector of each PE in the network has the same dimension as in the input space. Let the synaptic weight vector of PE i be denoted by Nimiiiiwwww],,,,[,,2,1 (2.11) where N is the total number of PEs in the network. Winning PE kx,2 Input Layer kx,1 kmx, Competition Layer

PAGE 28

16 To find the best match of the input vector kx with the synaptic weight vectors iw compare the Euclidean distance between kx and iw and select the smallest one as Niwxargminiikio,,1, (2.12) which sums up the essence of the competition process among the PEs. According to (2.12), is the subject of attention because we want the identity of PE i. The particular PE i that satisfies this condition is called the best-matching unit or winning PE for the input vector oi kx 2.2.2.2 Cooperative process The winning PE locates the center of a topological neighborhood of cooperating PEs. A topological neighborhood can be defined by many methods. In particular, a PE that is firing tends to excite the PEs in its immediate neighborhood more than those farther away from it, which is intuitively satisfying. This means that after classification of the input sample, the adaptation will be done not only for the winning PE but also for the neighbors of the PE which gives the best response. Let iio, denote the topological neighborhood function centered on the winning PE then a typical choice of is oi iio, 22,,2kiikiioorrexp (2.13) where oiirr represents the Euclidean distance in the output space between the i th PE and the winning PE and k is the effective width of the topological neighborhood. To satisfy the requirement that the size of the topological neighborhood shrinks with time, let the width of the topological neighborhood function decrease with time

PAGE 29

17 ,2,1,1kkexpok (2.14) where o is the value of at the initiation of the SOM algorithm, and 1 is a time constant. Thus, as time (i.e., the number of iterations) increases, the width decreases at an exponential rate, and the topological neighborhood shrinks in a corresponding manner. 2.2.2.3 Adaptive process The network can be trained with a simple Hebbian-like rule to train the weights of the winning PE and its neighbors. The neighboring PEs can be trained in proportion to their activity (Gaussian), or all neighbors within a certain distance can be trained equally. The learning rule can be described as follows: otherwisewiiwxwwkiokikkikkiki,),(,,,,1, (2.15) Notice that both the learning rate, k and neighborhood size, k are time dependent and are typically annealed (from large to small) to provide the best performance with the smallest training time. 2.2.3 Modeling Methodology Based on the SOM In this architecture of local linear modeling, the SOM is trained to position the local models in the embedded output space, Tdykkkkyyyy )1(1,,,, The SOM preserves topological relationships in the input space in such a way that neighboring inputs are mapped to neighboring PEs in the map space. Then, when each PE is extended with a local model it can actually learn the mapping )(,1kykfy in a supervised way. Each PE has an associated local model { iiba, } in (2.3) that represents the approximation of the local dynamics.

PAGE 30

18 The local model weights { iiba, } are computed directly from the desired signal samples and the input samples by a least square fit within a Voronoi region centered at the current winning PE chosen from jiy, y The size of the data samples in the region must be at least equal to the dy-dimensional basis vector. The design procedure for this local model is as follows: 1. Apply training data to the SOM and find the winning PE corresponding to the input y such that we have winner-input pairs. 2. Use the least square fit to find the local linear model coefficients for the winning PE, where desired output vector as oi Mjiy, MjforbayjiyiTijioooo1],,,[ (2.16) where ][ooiTiba is the sought linear model coefficients, M is the size of inputs involved in the winning PE oi Specifically, the least-squares problem XY (2.17) is solved for where is defined as a matrix that contains each input vector associated with the winning PE, and is defined as a vector that contains the target outputs. It is well-known that although the least-squares solution obtained from (2.17) is reasonably good when the noise level is low, the estimates tend to be biased for higher levels of noise. Addition of a single sample to a cluster can radically change the distances. Besides, the models will perform very well for that particular training set with very low error because it has memorized the training examples but they may not perform well with new data sets. Thus we make use of data samples from the winner as well as the neighbors to create the local models in order to make them more robust as well as to )(MdyX MY

PAGE 31

19 improve the generalization for the network. Also we take the data samples from the neighbors in case less data than the dimension of the input are assigned in some Voronoi region. In testing, once the winning PE is determined we select the appropriate local model from the list of associated models. Apply the local model to obtain the estimated output oikyTikbay,1 (2.18) 2.3 Input-Output Representation of Systems The temporal state evolution of an autonomous system is functionally dependent only on the system state, but a nonautonomous system, such as considered in this work, allows for an explicit dependence on an independent variable, the control input, in addition to the system state. For an autonomous system, it is reasonable to assume that the future behavior of the system can be predicted over some finite interval from a finite number of observations of past outputs. In contrast, predictions of the behavior of a nonautonomous system require consideration of not only the internal deterministic dynamics (past outputs), but also of the external driving term (future input) [30,75,96]. System identification is a technique that permits building mathematical models of dynamic systems based on input-output data (measurements). Its main purpose is to identify a model of an unknown process in order to predict and gain insight into the behavior of the process [39]. Real-life systems almost always show nonlinear dynamical behavior. This behavior complicates the task of finding models that accurately describe these systems. While in a large number of applications a linear model shows already satisfactory results, there are numerous situations where linear models are not accurate enough; especially when we deal with very complex systems or require very high

PAGE 32

20 performance. Physical knowledge of the system can be a great aid in finding a nonlinear model. However, this knowledge is not always available. In these cases we have to determine a model from a finite number of measurements of the systems inputs and outputs. This approach to nonlinear system modeling is often referred to as nonlinear black-box identification. Usually, a nonlinear mapping is fitted from a number of delayed inputs and outputs to the current output [94]. This results in a nonlinear input-output model of the system. 2.3.1 Classical Approach Some common classical approaches for nonlinear nonautonomous system modeling are based on polynomials, e.g., Kolmogorov-Gabor polynomial models [71], Volterra Series models [110], Hammerstein models [16,22,33], and so on, for the realization of the nonlinear mapping. ku 1z 1z 1z 1z 1z Figure 2-4. Nonlinear dynamic model configuration. Normally, a discrete-time nonlinear dynamic system can be described by a NARX (Nonlinear Auto-Regressive with eXogenous input) model that is an extension of the linear ARX model, and represents the system by a nonlinear mapping of past inputs and output terms to future outputs, that is, ),,,,,(111 dukkdykkkuuyyfy (2.19) ky Nonlinear static approximator )( f 1z 1ky

PAGE 33

21 Here is the output vector and is the input vector. For simplicity, we will set pkYy qkUu 1 qp Let the )(dudy dimensional basis vector be TdukkdykkTkukykuuyy],,,,,[],[11,, (2.20) where k is in the set Figure 2-4 shows the schematic diagram of NARX model. Another nonlinear model is a NOE (Nonlinear Output Error) structure described by dudyUY ),,,,,(111 dukkdykkkuuyyfy (2.21) where is the output of the identification model A NARMAX (Nonlinear Auto-Regressive Moving Average with eXogenous input) is the lagged version of the NARX model and is represented by y f kdukkdykkkeuuyyfy ),,,,,(111 (2.22) In the above, can be replaced by neural networks, radial basis function networks or fuzzy logic systems, which are other methods that have been developed for nonlinear system identification [5,96]. Narendra and Parthasarathy [65] have compared NOE and NARX, and as a result they have shown that NARX is better than NOE. In the neural network community most identification schemes use the series-parallel model (NARX). f 2.3.2 Series-Parallel and Parallel Models A nonlinear dynamic model can be used in two configurations: a series-parallel model and a parallel model. A series-parallel model predicts one or several steps into the future on the basis of previous plant inputs and plant outputs and ensures that all the signals are bounded if the plant is BIBO stable. Most published reports use the series-parallel model because of its resulting stability. A requirement for using this model is that

PAGE 34

22 the plant output is measured during the operation. In particular, in control engineering applications the series-parallel model plays an important role, e.g., for the design of a minimum variance or a predictive controller. Figure 2-5. A series-parallel model (left) and A parallel model (right). In contrast, a parallel model is required whenever the plant output cannot be measured during operation. This is the case when a plant is to be simulated without coupling to the real system, or when a sensor is to be replaced by a model. Also, for fault detection and diagnosis the plant output may be compared with the simulated model output in order to extract information from the residuals. Finally parallel model is very useful when dealing with noisy systems since it avoids problems of bias caused by noise on the real system output: If the identification model is to be used offline, the parallel model is obviously more suitable. The parallel model, however, lacks theoretical verification; hence, it is difficult to utilize its advantages. The two configurations shown in Figure 2-5 can not only be distinguished for the model operation phase but also during training. In this research, we follow the series-parallel model. ky 1ky ku )(,fModel )(,zPPlant 1z 1z 1z 1z 1z ky ku )(,zPPlant 1z 1z 1 1z z )(,fModel 1ky

PAGE 35

23 2.3.3 SOM-based Multiple ARX Models In the interest of modeling the local dynamics of a nonautonomous system in each region, the local approximation method presented for autonomous systems can be extended by letting kkx in (2.3), so that )(1kikfy Provided that necessary smoothness conditions on Yfi : are satisfied, a Taylor series expansion can be used around the operating point. The first-order approximation about the systems equilibrium point produces N local predictive ARX models of the plant described by Nff,,1 Niubyafdujjkjidyjjkjiki,,1,)(10,10, (2.23) where and are the parameters of the i jia, jib, th model. Although higher order Taylor approximations would improve accuracy, they are not very useful in practice because the number of parameters in the model increases drastically with the expansion order. Our proposed methodology is summarized as follows: first, the delayed version of input-output joint space is decomposed into a set of operating regimes that are assumed to cover the full operating space 1 Next, for each operating regime we choose a simple linear ARX model to capture the dynamics of the region. Consequently, a nonlinear nonautonomous system is approximated by a concatenation of local linear models ),()(,,1uyiNiff (2.24) 2.3.3.1 Selection of operating regions with a SOM Building local mappings in the full operating space is a time and memory consuming process, which led to the natural idea of quantizing the operating regimes and building local mappings in positions given by prototype vectors obtained from running

PAGE 36

24 the plant. For quantization of the operating regimes, the k-nearest-neighbor method is effective but it disregards neighborhood relations, which may affect performance [53]. In contrast, the SOM has the characteristic of being a local framework liable to limit the interference phenomenon and to preserve the topology of the data using neighborhood links between PEs. Neighboring PEs in the network compete with each other by means of mutual lateral interactions, and develop adaptively into specific detectors of different signal patterns [44]. The training algorithm is simple, robust to missing values, and it is easy to visualize the map. These properties make SOM a prominent tool in data mining [94]. In most of the papers discussing local linear models for system identification, the SOM has been used with a first order expansion around each PE in the output space. The SOM transforms an incoming signal pattern of arbitrary dimension into a one or two-dimensional discrete map, and performs this transformation adaptively in a topologically ordered fashion [44]. The results obtained so far with this methodology have been quite promising. However, problems that need to be solved remain: first, efficiently partitioning the operating regimes in high dimensional spaces is still a problem due to the curse of dimensionality [30]; second, it may be hard to find a small number of variables to characterize the operating regimes due to the possibly large number of local models; third, all the methods have to be extended for nonautonomous regimes. The previous work by Principe et al. [75] provided the starting point for the proposed modeling architecture. The most important difference is how to capture the dynamics in the input-output joint space, which is fundamental for identifying the unknown system. Several options are possible, and we have been investigating them:

PAGE 37

25 Firstly, we tried to find the local models by quantizing the input-output joint space by embedding not only the outputs but also the control inputs using one SOM. This modification is essential because the purpose is to characterize the system dynamics that exist in the input-output joint space. However, we encountered some difficulties such as normalization of the joint space and large dimensionality of the space involved (many degrees of freedom and large dynamic range of parameters) [13]. Secondly, in order to reduce the approximation error with local models based on a SOM, we utilized a counter-propagation network by quantizing the input-output joint space and the desired signal space together [15]. Since the output at each PE is just the average output for all of the feature vectors that map to that point local models might be created for better approximation using the quantization error in the input space and the average output. This is achieved by coupling each PE with a linear mapping in such a way that a functional relationship can be established between each Voronoi region in the input space (of the SOM) and the desired signal. However, this method required a much larger map to make the estimation error in the desired output space smaller. Additionally, when noise is added in the input of the SOM, the quantization error in the input Voronoi region may be magnified by the local models. As the number of dependent variables is increased, the process becomes increasingly difficult to model accurately. This led us to think that a model that uses only a few of the observed variables will be more accurate than a model that uses all the observed variables. In this scheme, therefore, we let the SOM look at only the current output and its past values to decide the winner, and create the models with the control inputs.

PAGE 38

26 Here we will pursue the last option for the following reasons. The competitive learning rule works best for normalized inputs. The SOM algorithm uses the Euclidean metric to measure distances between vectors. For example, if one variable has values in the range of [-100,,100] and another in the range of [-1,,1] the former almost completely dominates the map organization because of its greater impact on the measured distances. Either, the measure of distance is weighted by the inverse of the scales or the data must be normalized such that each component of the input vectors have unit variance and zero means [8]. However, normalization loses information (the mean or the scale can be important) and it can become meaningless if the data dynamic range (or mean) changes over time. Therefore we cannot normalize the data (nor create the weighted Euclidean metric) in this way since it is not always guaranteed that the mean and the dynamics range of the data are available. In addition, as the number of dependent variables is increased, SOM modeling becomes increasingly difficult because it is basically a memory-based approach that does not scale up well with the input dimension. This led us to think that a model that uses only a few of the observed variables will be more accurate than a model that uses all of the observed variables. When the SOM modeling is done in the output space, we let the SOM look at only the current output and its past values to decide the winner which represents the operating regime, and create the models with the control inputs as shown in Figure 2-6. In so doing, normalization of the input space is not necessary since the clustering is performed solely by the history of the output.

PAGE 39

27 Figure 2-6. Configuration of local linear modeling based on a SOM. 2.3.3.2 Model development procedure After the operating regions are divided by the SOM the underlying dynamics f is then approximated as where N is the number of operating regions. N local predictive ARX models of the plant are described by iNiff1 Nff,,1 Niubyafdujjkjidyjjkjiki,,1,)(0,0, (2.25) where and are the parameters of the i jia, jib, th model. Then, when each PE of the SOM is extended with a local model it can actually learn the mapping ),(,,1kukyikfy in a supervised way. The development of local models is done by directly fitting the quantized embedded output samples obtained from the SOM and corresponding embedded control input samples that cover the whole range of operation of the plant. ku, ky, ku SOM Embedding Embedding Plant ky 1a 1b 2a 2b Na Nb 1 0 1ky

PAGE 40

28 Each PE has an associated local model { iiba } which are computed directly from the desired signal samples and the input-output samples by a least square fit within a Voronoi region centered at the current winning PE chosen from jir, y The design procedure for this local model is as follows: 1. Apply training data to the SOM and find the winning PE corresponding to the input y such that we have winner-input pairs. 2. Use the least square fit to find the local linear model coefficients for the winning PE, where desired output vector as oi Mjior, MjforbarjiujiyTiTijiooooo,,,,, (2.26) where is the sought linear model coefficients, M ][TiTiooba is the size of data involved in the winning PE oi 3. In testing, once the winning PE is determined we select the appropriate local model from the list of associated models. Apply the local model to obtain the estimated output kuTikyTikoobay,,1 (2.27) Our proposed modeling methodology is summarized as follows: first, the delayed version of input-output joint space is decomposed into a set of operating regions that are assumed to cover the full operating space. Next, for each operating region we choose a simple linear ARX model to capture the dynamics of the region. Consequently, a nonlinear nonautonomous system is approximated by a concatenation of local linear models.

PAGE 41

CHAPTER 3 MULTIPLE MODEL BASED CONTROL Researchers have been interested in control of nonlinear systems for a very long time. Progress in nonlinear control design, however, has been difficult because of the intrinsic complexity of the problem [82]. In general, nonlinear control methods are complex and can be applied only to a narrow class of systems. For example, methods such as backstepping and feedback linearization can be applied to nonlinear systems with some specific structure, but not to arbitrary nonlinear systems. Thus, nonlinear control methods cannot serve all needs of real industrial control problems. One way to approach the control of a nonlinear system in a wide range of conditions is to linearize the model at a number of operating points, and then design one linear feedback controller at each operating region [101]. These local controllers can then be switched or scheduled as the system changes operating conditions. The use of multiple models is not novel in control theory. Multiple Kalman filters were proposed in the 1960s and 1970s by Magill [51] and Lainiotis [45] to improve the accuracy of the state estimates in control problems. Fault detection and control in aircraft was proposed by Maybeck and Pogoda [54], and in the subsequent years Maybeck and Stevens [55] used the idea extensively in controlling aircraft systems. In all the above cases, no switching is concerned, and only a linear combination of the control determined by the different models is used to control the system. The idea of switching between controllers has been most likely introduced for the first time in the adaptive control literature by Martensson [52]. In the direct switching 29

PAGE 42

30 schemes, the sequence in which the different controllers are to be tried is pre-determined. The only determination that has to be made is when to switch from one controller to another. It was soon realized that such architectures have very little practical utility. On the other hand, the outputs of the multiple observers determine both when and to which controller switching should occur in indirect switching schemes. Middleton et al. [56] explicitly proposed the use of multiple models and switching to alleviate the problem of stabilizing of the estimated model in indirect control, and further extended in [59] and labeled the Hysteresis switching algorithm. The objective in all the above efforts is to attain stability in adaptive control with minimum past information. A controller is often highly dependent on a plant model especially when the controller has been designed out of the model. Hence, for those cases, the modeling error would be a relevant criterion for controller selection. If the number of controllers is bounded, the delay between the selection of the controller and its activation can be neglected. Thus the selection of the controller according to the modeling error is feasible. Narendra and Balakrishnan [64] were the first to propose this idea of using multiple adaptive models and switching in order to improve the performance of an adaptive system, while assuring stability. Although it has already been shown that the performance of a system can be significantly improved using the multiple model adaptive control with switching, applicability to highly complex systems with this approach has not been investigated in details. Thus, in this chapter, multiple controller design methodologies are introduced by extending the multiple model approach for more complex nonlinear systems.

PAGE 43

31 3.1 Discrete-Time Control System The field of classical control theory concerns itself with the task of servo or regulator control of linear analog plant. Design methods for both continuous-time linear controllers and discrete-time linear controllers obtained by discretizing the plant are well understood. Figure 3-1 shows a schematic diagram of a classical discrete-time control system. kd Figure 3-1. Classical discrete-time control system. The signal is the reference signal. We would like the plant output to track it as closely as possible. To track the reference signal, the controller uses both and to compute the plant control signal Feedback of is used to stabilize the plant, and to ensure that the controller is both resilient in the face of external disturbances and able to quickly reduce the output error to zero. Their only drawback is that they assume precise knowledge of the plant dynamics. For this reason, a great deal of effort has been expended to create accurate models of typical plants. As one improved way of modeling, we proposed multiple models for better approximation of the plant in Chapter 2. This scheme makes it easier to design the controller for the model which approximates internal descriptions of the plant when given a finite number of external measurements without any knowledge of the plant dynamics. 1kr 1ky 1kr ky ku ky ku 1ky Controller Plant ) ( z P 1kr ) ( z G 1z

PAGE 44

32 One reason to consider discrete-time systems is that it is well known that most complex systems are controlled by computers which are discrete in nature and this constitutes an obvious reason for dealing with multiple model based control. Another is the fact that the presence of random noise can be dealt with more easily in the case of discrete-time systems. Since most practical systems have to operate in the presence of noise, the stability and performance of multiple model based control in such contexts has to be well understood, if the theory is to find wide applications in practice. 3.2 Inverse Control via Backpropagation Through Model In order to design a controller, we need to determine a plant model which should capture the dynamics of the plant well enough that a controller designed to control the plant model will also control the plant very well. Such a model might be derived from physics by carefully analyzing the system and determining a set of partial differential equations which explain its dynamics. Alternatively, the model might be a black-box implementing some sort of universal transfer function. This function may be tuned by the adjustment of its internal parameters to capture the dynamics of the system. For nonlinear unknown systems, a NARX model of sufficient order is a universal dynamic system approximator. Hence, we implement NARX neural network plant models for performance comparisons with the proposed local linear models. The conventional design methods for control systems involve constructing a mathematical model of the systems dynamics and utilization of analytical techniques for the derivation of a control law. Such mathematical models comprise sets of linear or nonlinear differential/difference equations, which are usually derived with a degree of simplification or approximation. The modeling of physical systems for feedback control generally involves a balance between model accuracy and model simplicity [102]. Should

PAGE 45

33 a representative mathematical model be difficult to obtain, due to uncertainty or sheer complexity, conventional techniques prove to be less useful. Also, even though an accurate model may be produced, the underlying nature of the model may make its utilization using conventional control design difficult. Neural networks, hence, have been used for different purposes in the context of control due to its ability to learn an essential feature of unknown plants by mimic [43,105,78]. Most of methods existing are based on inverse control. We will therefore start by considering a neural network controller, specifically, TDNN since this is very suitable to create both a model and a controller when only input-output measurements are available. Principally, the TDNN is an extended multilayer perceptron that allows us to handle temporal patterns and the problems of time variant signals, i.e., signals that are scaled and translated over time [40]. The idea that has been followed in the TDNN is based on the invention of time delays, resulting in giving the individual PEs the ability to store the history of their input signals. This way, the network as a whole can adapt not only to a set of patterns, but also to a set of sequences of patterns. An advantage of the TDNN is the relatively simple mathematical analysis and ability of training by Backpropagation algorithm. Thus we compare the performance of the proposed control systems with that of an inverse controller trained through the TDNN. The algorithm is derived as follows: First we train a TDNN as a model, by letting as shown in Figure 3-2(a). Then a TDNN controller is designed based on the created TDNN model from which we obtain the Jacobian of the plant. The controller is described by f ),,,,,(111kkkkkuuyyfy

PAGE 46

34 ),,,,,,,(1121Wyyruugukkkkkk letting g be the function implemented by the controller Here W is the weights of the TDNN model, The controller parameters in the fixed control structure are adapted by an algorithm that ensures that the desired performance level is maintained and the parameters are updated by back propagating the error through the model as shown in Figure 3-2(b). )(zG )(zF (a) (b) Figure 3-2. Modeling and control scheme using the TDNN: (a) TDNN modeling of a plant (b) An inverse controller via Backpropagation through (Plant) Model. 3.3 Multiple Inverse Control Now we discuss the control problem for the local linear model using an inverse control framework [17]. The central advantage of such a framework is that an inverse model can be used directly to build a feed-forward controller. When given a model the control network is brought on line and the control signal is calculated at each instant of time by setting the output value at instance ) ,,,(111kkkkikuuyyfy 1ky 1 k equal to the desired value as while trained off line as in classical inverse control approach. Thus, for the desired behavior, the controller simply asks the model to predict the action needed. 1kr ),,,(1111kkkkikuryyfu )(zP Embedding )(zF ku ky )(zG 1kr 1z 1kr )(zP ku Embedding )(zF ky1ky

PAGE 47

35 As stated before, our principal objective is to determine a control input, which will result in the output, of the plant tracking with sufficient accuracy a specified sequence, The system identification block has N predictive models denoted by in parallel. Corresponding to each model a controller is designed such that achieves the control objective for Therefore, at every instant one of the models is selected and the corresponding controller is used to control the actual plant. In order to control a plant, consider the control problem where the dimension of the input is equal to that of the output, that is, ku 1ky 1kr Niif1 if ig ig if q p From (2.23), because q p and under the assumption that is invertible, the control law of an inverse controller for the model, can be directly calculated as ob oif 11,10,110,dujjkjidyjjkjikikubyarbuooo (3.1) Therefore, at time instance k, the control can be obtained, if the future target of , is known. Therefore, the set of local linear models simplifies the control design for a nonlinear plant. So instead of a global neuro-controller as in other adaptive control schemes [12,18,31,58], here we can function with a group of linear controllers associated with each identified model, thus taking care of the system over the whole operating region. ku ky 1kr One advantage of this scheme is its simplicity and fast convergence to get the desired response. Another advantage is that the dynamic space is decomposed in the appropriate switching among very simple linear models, which leads to accurate modeling and controls. On the other hand, creating a set of models by embedded input

PAGE 48

36 and output may cause serious problem in the presence of large noise or outliers since the wrong predictive model due to noise may cause poor control. Hence, the selection of the right model is as important as creating models and designing controllers. Once the right local linear model is determined, the corresponding controller is designed using (3.1). A schematic diagram of the proposed SOM based inverse control system is shown in Figure 3-3 where the inverse control seeks to model the inverse of the plant. A set of controllers appears in series with the plant. The command input, is fed to the controller and provides also the desired response. Hence, when the error is small the controller transfer function is the inverse of the plant. 1kr ku ky Figure 3-3. Proposed SOM-based inverse control scheme. Generally, an adaptive controller that meets the specifications is slow to adapt. However, our approach models all the operating regions and automatically divides the operating regions by the number of PEs. So once the current operating region is determined by the SOM, the corresponding controller is triggered so that the plant tracks the desired signal. Moreover, even if the wrong PE is assigned in the winning PE due to noise, a similar dynamic model can be activated since neighboring SOM PEs represent 1kr SOM Embedding Embedding 1f 2f Plant )(zP 111gf 212gf Nf NNgf1

PAGE 49

37 neighboring regions in the dynamic space. Thus, the proposed control system can reach the set point quickly, and even if the dynamic model is not the most appropriate, there is an extra flexibility to match the set point with the least amount of error. 3.4 Multiple PID Control In linear control theory, despite the development of more sophisticated control strategies, Proportional-Integral-Derivative (PID) controllers have been extensively studied by researchers and well understood by practitioners, since they are widely used in practice, and their principle is well understood by engineers [95,92]. It gained its popularity for its simplicity of having only three parameters. But owing to its simplicity it has also paid the price of not having an efficient and practical way of determining optimal gains. The ideal continuous time PID controller is expressed in Laplace form as follows: sKsKKsGdip)( (3.2) where is the proportional gain, the integral gain, and the derivative gain. Each of the terms works independently of the other pK iK dK 1 The standard PID control configuration is shown in Figure 3-4. The introduction of integral action facilitates the achievement of equality between the measured value and the desired value, as a constant error produces an increasing controller output. The derivative action indicates that changes in the desired value may be anticipated, and thus an appropriate correction may be added prior to the actual change. Thus, in simplified terms, the PID controller allows contributions from present, past and future controller inputs. 1 This is not exactly true since the whole thing operates in the context of a closed-loop. However, at any instant in time, this is true and makes working with the PID controller much easier than other controller designs.

PAGE 50

38 )(sG Figure 3-4. PID controlled system. Extensions of the PID control methodology to nonlinear systems, however, are not trivial. Usually, global nonlinear models are, as always, linearized and the parameters of the PID controller are scheduled according to the regime. All these difficulties, in fact stem from the fact that the system model is constructed from a nonlinear dynamical equation. This difficulty could be eliminated by a piece-wise linear approximation of the nonlinear dynamics (as one does in gain scheduling). However, gain scheduling methods are not flexible for model uncertainties and they cannot be scaled up to regimes that are not described in the initial system identification stage by a linear model. It also has the problem of either inefficient use of system identification data due to a large number of arbitrarily selected operating (linearization) points or inaccurate modeling due to the small number of linearization points [83]. The multiple PID control design method described in this section is rooted in the principle of using local linear models to construct a globally nonlinear (piecewise linear) system model that is determined completely from the input-output data collected from the actual plant [12,30]. In addition to previously designed multiple inverse control schemes, this section illustrates how the multiple models can be united with the well-known linear PID controller design techniques to obtain a principled and simple nonlinear PID )(sY )(sU )(sR )(sE pK sKi sKd )(sP

PAGE 51

39 controller design methodology. The proposed closed loop scheme is illustrated in Figure 3-5 where the SOM determines which linear PID controller contributes to the instantaneous control input. Figure 3-5. Overall schematic diagram of the nonlinear PID closed loop control mechanism using multiple controllers. Once system identification is complete, the design of a globally piece-wise linear PID control system can be easily accomplished using standard techniques. The literature has an abundance of PID design methodologies for linear SISO systems including direct pole-placement techniques and optimal coefficient adjustment according to some criteria [95,92]. Here pole-placement technique [4] is utilized to design a PID controller for each linear SISO model and is illustrated briefly in the following. Assume that the plant is modeled by a set of the SISO, second-order system given by 1,2,11,2,1111,,, kikikikikkkkikububyayauuyyfy (3.3) where is the output of the plant, and is the input to the plant. The controller is designed by starting with a general PID regulator shown in Figure 3-6 and determining ky ku kd 1kr 1ky ku )(1zG )(2zG )(zGN SOM )(zP Embedding

PAGE 52

40 the coefficients of polynomials and so that the closed-loop system has desired properties. iC iD Figure 3-6. Block diagram of PID controller for a SISO plant model. Each models transfer function is given by iiiiiiiazazzbzbzAzBzF,2,12,22,1)()()( (3.4) It is assumed that the polynomials and do not have any common factors. Note that in order to have a stable zero (zero inside the unit circle), must be true. The controller specifications are expressed in terms of a model that gives the desired response to command signals. The general control equation is )(zAi )(zBi iibb,2,1 )()()()()(zYzRzDzUzCii (3.5) It is assumed that the is monic. To make sure that low-frequency disturbances give small errors, is chosen as iC iC )()1()('zCzzCiLi (3.6) with a suitable selection of This puts a L )1( z term in the denominator of the controller transfer function, guaranteeing integral control action. The goal of the controller design is to map the values of and into the controller coefficients, and subject to the constraint of the model whose ,,,,1,2,1iiibaa ib,2 ijd, ic )(zU )1( zY )(zE )1(zR )()()(zCzDzGiii )()()(zAzBzFiii 1z

PAGE 53

41 characteristic polynomial is The transfer function of the closed-loop system is 212zz )()(/)()(11zDzBzCzzAiiii ; therefore, its characteristic polynomial is The characteristic polynomial for the model is with the addition of a second-order observer polynomial term Equating characteristic polynomials results in the design equation )()()()(1zDzBzzCzAiiii mA oA )()()()()()1)((1'zAzAzDzBzzCzzAmoiiiLi (3.7) which can be written as )())(())(1)((2122,3,22,1,22,11,2,12zzzdzdzdzbzbzczzazaziiiiiiii (3.8) This equation is solved for the and Then the control law difference equation is ijd, ic 2,31,2,121)1( kikikikikikedededucucu (3.9) Given the local linear models as obtained through the use of a SOM and a PID design technique, the overall closed loop nonlinear PID design reduces to determining the coefficients of the individual local linear PID controllers using their respective linear plant model transfer functions. One needs to determine a set of PID coefficients per linear model. In the competitive SOM approach, the model output depends only on a single linear model at a given time; therefore, the PID coefficients are set to those values determined for the instantaneous winning model.

PAGE 54

CHAPTER 4 MULTIPLE QUASI-SLIDING MODE CONTROL Attempts to use traditional control methods, such as inverse control and PID control, for nonlinear plants will inevitably encounter problems when faced with the nonlinear nature of these systems. In order to overcome these difficulties in designing controllers for nonlinear systems, a simplified control method that keeps the advantage of the conventional approach was proposed in Chapter 3, i.e., the SOM was explored as a modeling infrastructure, and the controllers were built based on SOM-based multiple models. However, these control methods (especially multiple inverse control scheme) may show poor control performance when existing sensor noise or external disturbances due to the fact that the perfect control is achieved if the plant and the controller is stable, if the model is perfect, and if there is no disturbance. Wrong selection of the model caused by noise can devastate the control mission since the controller is also determined by very different model from the given task. For this reason, sliding mode control architecture, which is a very well-known robust controller, is employed in this chapter in order to obtain sturdiness against noise as well as uncertainties on the model. 4.1 Introduction to Variable Structure Systems The control of nonlinear systems has been an important research topic and many approaches have been proposed. While classical control techniques have produced many highly reliable and effective control systems, great attention has been devoted to the design of variable structure control systems (VSCS). Variable structure systems (VSS) 42

PAGE 55

43 are a special class of nonlinear systems characterized by a discontinuous control action which changes structure upon reaching a set of switching hyperplanes during the control process to attain improved overall characteristics in the controlled system. During the sliding mode the VSCS has invariance properties, yielding motion that is remarkably good in rejecting certain disturbances and parameter variations [93,9,25,86]. Most of the VSCS proposed in the literature have been developed mainly based on the state-space model with the assumption that all state variables are measurable or on the input-output model for a linear system. But in some control problems, we are allowed to access only the input and the output of the nonlinear plant [72]. In this case, an observer could be used to estimate the unmeasurable state variables if the state equations are known. Otherwise, this is not possible. Thus, it is the purpose of this chapter to provide a new technique to design sliding mode control systems based on input-output models of the considered discrete-time nonlinear system so that the amount of guesswork 1 is reduced, while attainable performance is increased. Normally, the design of VSS consists of two parts: First, the sliding surface, which is usually of lower order than the given process, must be constructed such that the system performance during sliding mode satisfies the design objectives, in terms of stability, performance index minimization, linearization of nonlinearities, order reduction, etc. Second, the switched feedback control is designed such that it satisfies the reaching condition and thus drives the state trajectory to the sliding surface in finite time and maintains it there thereafter [93]. 1 In most cases, we need to estimate the unknown parameters, unmodeled dynamics and bounded disturbances. Also, it should be noted that the VSCS works best when the plant is completely known.

PAGE 56

44 4.1.1 Sliding Hyperplane Design Sliding surfaces can be either linear or nonlinear. The theory of designing linear switching surfaces for linear dynamic system has been developed in great depth and completeness. While the design of sliding surfaces for more general nonlinear systems remains a largely open problem. For simplicity, we focus only on linear switching surfaces. Moreover, for surface design, it is sufficient to consider only ideal systems, i.e., without uncertainties and disturbances. Consider a general system uxBxAx)()( (4.1) with a sliding surface 0)(| xSxS where are general nonlinear functions of )(xA )(xB x and nx mu The equivalent control is found by recognizing that is a necessary condition for the state trajectory to stay on the sliding surface Therefore, setting i.e., 0)(xS 0)(xS 0)(xS 0)()()(equxBxSxAxSxxSxS (4.2) yields the equivalent control )()(1xAxSxBxSueq (4.3) where is nonsingular. When in sliding mode, the dynamics of the system is governed by )()/(xBxS )()()(1xAxSxBxSxBIx (4.4)

PAGE 57

45 For example, if the system (4.1) is linear and described by BuAxx (4.5) where A and B are properly dimensioned constant matrices. The switching surface can be defined as 0)(xCxST (4.6) i.e., where TCxS/ TmcccC,,,21 is an nm matrix, and then we have AxCBCuTTeq1)( (4.7) and (4.4) becomes xBKAAxCBCBIxTT)())((1 (4.8) (4.4) and (4.8) describe the behavior of the systems (4.1) and (4.5), respectively, which are restricted to the switching surface if the initial condition satisfies )(0tx 0))((0 txS For the linear case, the system dynamics is ensured by a suitable choice of the feedback matrix In other words, the choice of the matrix can be made without prior knowledge of the form of the control u. ACBCKTT1)( C 4.1.2 Sliding Mode Control Law Design Once the sliding surfaces have been selected, attention must be turned to solving the reachability problem. This involves the selection of a state feedback control function which can drive the state mnu: x towards the surface and thereafter maintains it on the surface illustrating in Figure 4-1. In other words, the controlled system must satisfy the reaching conditions. The classic sufficient condition for sliding mode to appear is to satisfy the condition and a similar condition proposed by Utkin [93], i.e., missii,,1,0

PAGE 58

46 0lim0issi and These reaching laws result in a VSC where individual switching surfaces and their intersection are all sliding surfaces. This reaching is global but does not guarantee finite reaching time. 0lim0issi x 0)( xsi x Reaching phase Sliding phase Reaching phase Figure 4-1. Phase plane plot of a continuous-time second-order variable structure system. Another commonly used reaching law is proposed by Gao & Hung [27]. The law directly specifies the dynamics of the switching surface by the differential equation )()sgn(SRgSQS (4.9) where the gains and Q R are diagonal matrices with positive elements, and TmssS)sgn(,),sgn()sgn(1 TmmsgsgSg)(,),()(11 where 0)(1,0)(0,0)(1)sgn( xsifxsifxsifsiiii (4.10) and the scalar functions satisfy the condition ig 0)(iiisgs when 0 is (4.11) Various choices of Q and R specify different rates for approaching and yield different structures in the reaching law. S

PAGE 59

47 For one way of designing controllers, recall that during sliding mode, one can compute the equivalent control according to (4.3) or (4.7). However, only using cannot drive the state towards the sliding surface if the initial conditions of the system are not on S. One popular design method is to augment the equivalent control with a discontinuous or switched part, i.e., equ equ Nequuu (4.12) where is a continuous control defined by (4.3), and is added to satisfy the reaching condition which may have different forms. For a controller having the structure of (4.12), we have equ Nu NNeqNequxB x SuxBxSuxBxAxSuuxBxAxSxxSxS)()()()())(()()( for simplicity, assume IxBxS )()/(, then Some often used forms of are relay type of control NuxS)( Nu ))(sgn(xSuN linear continuous feedback control ))((xSuN and linear feedback control with switching gains where xuN ][ij is an matrix and nm 0)(0)(jiijjiijijxxsifxxsif Parameters ij and ij are chosen to satisfy the desired reaching condition.

PAGE 60

48 Another design method of controllers is to employ the reaching law approach proposed by Gao & Hung [27] and can be directly obtained by computing along the reaching mode trajectory, i.e., )(xS ))(())(sgn()()()(xSRgxSQuxBxAxSxxSxS (4.13) Hence, we have ))(())(sgn()()(xSRgxSQxAxSxBxSuT (4.14) By this approach, the resulting sliding mode is not preassigned but follows the natural trajectory on a first-reach-first-switch scheme. The switching takes place depending on the location of the initial state. 4.2 Sliding Mode Control in Sampled-Data Systems The VSS theory which was originally developed from a continuous time perspective has been realized that directly applying the theory to discrete-time systems will confront some unconquerable problems, such as the limited sampling frequency, sample/hold effects and discretization errors [2]. Since the switching frequency in sampled-data systems cannot exceed the sampling frequency, a discontinuous control does not enable generation of motion in an arbitrary manifold in discrete-time systems [3,9]. This leads to chattering along the designed sliding surface, or even instability in case of a too large switching gain [2]. Figure 4-2 illustrates that in discrete-time systems, the state moves around the sliding surface in a zigzag manner at the sampling frequency. Much research has been done in this field. Among various concepts of discrete-time sliding mode, Quasi-Sliding Mode (QSM) is reviewed in this section for the sliding mode controller design aimed at sampled-data systems.

PAGE 61

49 Figure 4-2. Discrete-time system response with sliding mode control. 4.2.1 Quasi-Sliding Mode Many researchers have either addressed the limitations when direct implementation is done or have proposed designs that take the sampling process into account. Milosavljevic [57] was among the first researchers to formally state that the sampling process limits the existence of a true sliding mode. In light of this, the concept of QSM has been suggested and the conditions for the existence of such mode have been investigated. Consider a sampled-data SISO system with the predefined sliding surface 01kTkkkkxCsBuAxx (4.15) The desired state trajectory of a discrete-time VSC system should have the following features: Firstly, the trajectory moves monotonically towards the switching manifold and crosses it in finite time starting from any initial point. Secondly, the trajectory crosses the manifold in succession after it hits the manifold, resulting in a zigzag motion about the switching surface. Lastly, the trajectory keeps on within a specified band without increasing the size of each successive zigzagging step. Gao et al. [28] defined a QSM as the motion of a discrete VSC system satisfying last two features. In addition they named the specified band, )(|xsx which contains the QSM as the Quasi-Sliding Mode Band (QSMB). For single input systems, the main k+2 k+1 k State trajectory Sliding surface 0 s

PAGE 62

50 approaches for the design of QSM control laws can be categorized into the following two methods: Discrete Lyapunov function based approach : Sarpturk et al. [81] noticed that unlike the case in continuous-time SMC, the switching control in the discrete-time case should be both upper and lower bounded in an open interval, in order to guarantee the convergence of sliding mode. Recall that in continuous-time SMC, the control (4.12) is composed of the equivalent control and a switching control. Converting this control to discrete-time gives Nkeqkkuuu,, (4.16) Hui & Zak [36] observed that if is a relay control with a constant amplitude, the relay must be turned off in some neighborhood of the surface, in order to reach the switching surface, otherwise, the trajectory will chatter around the surface with a chatter amplitude at least as large as the amplitude of the relay output. The idea of sliding sector [26,48] was used to solve this problem, i.e., to specify a region in a neighborhood along the sliding mode, where linear control is used to keep the state inside the region after it has reached the region. The switching control is applied only when system states are out of the region. In this case, the derived switching surface is different from the sliding surface. Based on a discrete Lyapunov function, Nku, 221kkSV the reaching law is given by 211)(21)(kkkkkSSSSS for 0 kS (4.17) which ensures Furuta [25] proposed a control law of the type kkVV1

PAGE 63

51 kDeqkkxFuu (4.18) where the equivalent control is the solution of eqku, 01 kkkSSS (4.19) and therefore the equivalent control for the system (4.16) is kTTeqkxIACBCu)()(1, (4.20) DF is a discontinuous control law which will be zero inside the sliding sector. Reaching law based approach : Gao et al. [28] pointed out that the reaching law (4.17) was incomplete for a satisfactory guarantee of a discrete-time sliding mode, since it does not ensure that the trajectory moves monotonically towards the switching surface and the trajectory stays on within a specified band. Thus, they presented an algorithm that drives the system state to the vicinity of a switching hyperplane in the state space, rather than to a sector of a different shape. They specified desired properties of the controlled systems and proposed a reaching law based approach for designing the discrete-time sliding mode control law. The equivalent form of the reaching law for discrete-time SMC extended from the continuous-time reaching law (4.13), and for a SISO system is kkkkrTssqTss )sgn(1 , 0r 0q 01 rT (4.21) where is the sampling period. The state reaches the switching surface at a constant rate and the term 0T qT r T forces the state to approach the switching surfaces faster when is large. The inequality for T guarantees that starting from any initial state, the trajectory will move monotonically towards the switching surface and cross it in finite time. Then the control law for discrete SMC is derived by comparing ks kTkTkTkTkTkkxCBuCAxCxCxCss11 (4.22)

PAGE 64

52 with the reaching law (4.21), which yields, kkkTkTTkrTssqTxCAxCBCu)sgn()(1 (4.23) 4.2.2 Quasi-Sliding Mode Control Using Multiple Models Now we discuss the design of the control law for local linear models using the QSM control framework proposed by Gao et al. [28], where the system states move in a neighborhood around the sliding surface 0 ks The central advantage of the sliding mode control strategy is that it is an effective robust control strategy for incompletely modeled or uncertain systems. Thus, the feature of the proposed control scheme is that the robustness for disturbances can be obtained by the simple control logic based on the linear model for each region. Another feature of the strategy is that it guarantees convergence of the system output to a vicinity of the predetermined, fixed plane in finite time, specified a priori by the designer. Consider one of the local single input-output models of the plant if f 112111211 nknkkmkmkkkubububyayayay (4.24) Equivalently, the input-output model of the plant in (4.24) can be written as the state-space model 2 11211 nknkkkkuuuxx (4.25) where mTkkmkkyyyx],,,[11 is the system state vector which is available for measurement and and n ,,1 have the following forms: 2 In a formal state-space model, past values of the input should be included in the state vector using delay operators. For simplicity, we include only the system outputs past values in the state vector in this notation.

PAGE 65

53 1321100000010000010aaaaammmm nnbbb000,,000,0002211 Also defining the tracking error vector as 111 kkkxre (4.26) where the desired signal vector is Tkkmkkdddr],,[1,21 the switching surface is defined in the space of the tracking error vector given by kTkecs (4.27) where Tmcccc],,,[21 Then an equivalent control is designed to satisfy the ideal quasi-sliding mode condition, 01 kkss by 112111)()(nknTkTkkTTeqkucucxrccu (4.28) and the closed-loop system response of the ideal quasi-sliding mode substituting (4.28) into with an equivalent control is given by 11111111)()(kTTkTTkrccxccIx (4.29) The system (4.29) can be viewed as a linear system with the input 1kr and the output 1kx To get an insight into the tracking capability of the system, (4.29) can be represented in terms of the tracking error kkkyde by

PAGE 66

54 211211mkmkmmkmmkecceccecce (4.30) Note that by designing the switching surface such that the roots of polynomial are inside of the unit circle, the error vanishes and thus the condition ensures asymptotic convergence to the desired output. An arbitrary positive scalar also determines the time taken to reach the sliding surface and can be adjusted to get a faster response. )/()/(1211mmmmmcccc mc The reaching law (4.21) always satisfies the reaching condition such that the discrete VSC system designed using the reaching law approach is always stable with a stable ideal quasi-sliding mode [28]. Then the control law is derived by comparing kTkTkkececss 11 with the reaching law (4.21), which yields, )sgn()1(12111kknknTkTkTkTTksqTsrTucucxcrccu (4.31) Salient feature of the multiple quasi-sliding mode controller is that one can obtain faster convergence to get the desired response due to multiple control scheme and one can employ VSCS to control unknown nonlinear plants while gaining indemnity against noise and parameter variations. 4.3 Analysis of Multiple Quasi-Sliding Mode Control with an Imperfect Sensor For the SOM-based system identification, one needs to quantify the effect of the modeling error that will occur due to the quantization of state space induced by the SOM, and also by the wrong selection of the winning model. Consider model (4.24) in the presence of modeling error and measurement noise. The predicted output becomes:

PAGE 67

55 11211111211)()()( nknkkmkmkmkkkkkubububyayayay (4.31) where a wrong local model nmbbbaaa,,,,,,,2121 is triggered by the SOM due to the noisy output measurement kky Then, when 0 ks the overall tracking error response with an equivalent control is given by 11221111211211)()())(())((nknnkmkmkmmkkmkmkmmkmmkubbubbyaayaaecceccecce (4.32) For simplicity, consider the error dynamics (4.32) when Defining model parameter error and noise 2m Tkbbaaaa],,[112211 Tkkkkn],,[21 )(1kkTkkknzece (4.33) where Tkkkkuyyz],,[11 21/ccc and c is chosen as 1c We assume that and ][2kE 22kkkww where the norm kkww is the Euclidean distance between the reference vector of the correct PE and that of the neighboring PE selected by the perturbed output measurements 3 Also, it is assumed that the noise k is zero-mean and white. We have the following recursive formula for the tracking mean squared error. kTkkTkkTkkTkkkTkkkkTkkkTkkkknnEzzeEcnznznzececEeE ][][]))(()(2[][222221 (4.34) When we take the norm on each side in (4.34), the norm of the tracking error power is represented by 3 The first assumption states that measurement noise has finite power. The second assumption means model parameter error is bounded by the distance in the state space.

PAGE 68

56 2222221][][kkkkkzeEceE (4.35) where the Cauchy inequality is used on the second term on the right hand side. As using the earlier assumptions on noise and model error bounds, the following bound on the steady-state tracking error power is obtained: k 2222)1(][kkkzcwweE (4.36) Note that the difference between the true model (winning PE) and the neighboring model (wrong PE) assigned by noisy input, 2kkww is typically small, since neighboring SOM PEs represent neighboring regions in the dynamic space. Also, it should be noted that the error can still be very large if we choose c as close as 1. In contrast, by choosing c as small as possible, the closed-loop system may have very fast transient response, possibly too large unexpected overshoot. Thus we should be careful for determining c so as not to have large error. This problem will be discussed later in simulation results. If c is set to small enough it then follows that the error by choosing appropriate design parameters mentioned above will be bounded for a given modeling uncertainty and measurement noise bounded by Moreover, this shows that the switching scheme does not create an issue to be considered in order to guarantee BIBO stability of the overall system.

PAGE 69

CHAPTER 5 CASE STUDIES To examine the effectiveness of the proposed multiple controller design methodologies, the chaotic systems with the input term, discrete-time systems, and flight vehicles have been considered assuming the following: Assumption 1: The only state available for measurements is kkxy,1 Assumption 2: The nonlinear function is completely unknown. f By assuming that the function is unknown, we confront a worst case (least prior knowledge) control design. If an estimate of is available, it can be included in the control design. As can be expected, the better the estimate the better the performance of the resulting controller. Our objective is to design multiple controllers for unknown nonlinear plants that guarantees global stability and forces the output, to asymptotically track the desired signal, i.e., f f f f ky 0kkry as without any a priori knowledge of the plant. k 5.1 Controlled Chaotic Systems Chaotic systems exhibit irregular, complex, and unpredictable behavior that exists in many industrial systems. Recently considerable effort has been in the focus of attention in the nonlinear dynamics literature since removing chaos can improve system performance, avoid fatigue failure of the system, and lead to a predictable system behavior. In the literature, several design techniques have been applied for the control and synchronization of a variety of chaotic systems. A generalized synchronization of 57

PAGE 70

58 chaos via linear transformation [106], adaptive control and synchronization of chaotic systems using Lyapunov theory [103,107], nonadaptive and adaptive control systems based on backstepping design techniques [31,32,99,100], and an adaptive variable structure control system for the tracking of periodic orbits [108] have been considered, to name a few. The contribution of this work lies in the design of multiple control system for the control of chaos based on the theory of the inverse control, PID control, and sliding mode control [14] assuming that the chaotic system is unknown. Given an unknown chaotic system, the goal is to force it to set points or a stable trajectory. 5.1.1 The Lorenz System The Lorenz model is used for fluid conviction that describes some feature of the atmospheric dynamic [73]. The controlled model is given by 321313122121xxxxuxxxxx)x(xx (5.1) where ,, and represent measurements of fluid velocity and horizontal and vertical temperature variations, respectively. The Lorenz system can exhibit quite complex dynamics depending on the parameter values. For 1x 2x 3x 10 the origin is a stable equilibrium point. For )1()3(:),(1* the system has two stable equilibrium points )1(,)1(,)1( and one unstable equilibrium point at the origin. For all three equilibrium points become unstable and the system trajectory have chaotic behavior [109]. As in the commonly studied case, we select ),(* 10 3/8 which leads to Thus we will consider the system with 74.24),(* 10 3/8 and 28 which produces the well-known butterfly

PAGE 71

59 chaotic dynamics without any control, i.e., 0 u as shown in Figure 5-1. Here, the control force, which is uniformly distributed in 100u is added into the second equation of (5.1). Figure 5-1. The uncontrolled Lorenz system: phase-space trajectory and time-series. After solving the set (5.1) with forth-order Runge-Kutta method with integral step 0.05, 7000 samples were generated for analysis. In order to construct a set of model-based controller, first, we design a model as 2111,,,, kkkkkkuuuyyfy (5.2) based on the Lipschitz index [35] as shown in Table 5.1. Table 5-1. Lipschitz index of the controlled Lorenz system for determining an embedding dimension. Number of inputs 1 2 3 4 5 1 1090.9 49.7 5.7 3.3 2.1 2 10.5 2.9 2.4 2.1 1.6 3 5.1 2.7 2.0 1.5 1.3 4 3.2 2.3 1.9 1.4 1.2 Number of outputs 5 3.0 1.8 1.4 1.4 1.2

PAGE 72

60 Figure 5-2. Generalization error v.s. Number of PEs (left) Learning curve (right). The SOM was trained with the output history vector Tkkkyyy],[1, over samples with the time decaying parameters, 6000L )003.01/(1.0kk and )003.01/()2/(kNk in (2.14), and then local linear models were built from ky, as well as Tkkkkuuuu],,[21, for each PE. A newly generated sequence of 1000 M samples was applied to multiple models for performance test 1 Then the dimension of the square SOM was determined as 8 based on the generalization error as shown in Figure 5-2 since the error does not decrease much after 8N Also, with 88 N the learning curve 2 for 4000 epochs is as shown in Figure 5-2. This curve reflects the overall closeness of the winning PE to the input samples during the training process and it becomes approximately constant at the end of training. 1 Identification performance was evaluated by NRMSE (Normalized root mean squared error): MkkkyrMr211)(1)max(/1 2 The learning curve is defined as the RMS (Root mean squared) distortion between the input and the winning PE: LkkikyowL2,,||1

PAGE 73

61 The identification result is shown in Figure 5-3 where the dashed line is the output of the controlled Lorenz system, and the solid line is the output from the multiple models. As we can see, the multiple models are a very good approximation of the controlled Lorenz system. In addition, plant modeling performance with 64 multiple models was compared with a single linear model, ARX, with the same number of inputs used in local modeling. Figure 5-3. Identification of the controlled Lorenz system by multiple models. It was also compared with those by means of a conventional TDNN, which was trained by the backpropagation algorithm with the constant learning rate of 0.001 for 3000 iterations on the same number of inputs and outputs as in local linear modeling, adopted as a global nonlinear model and listed in Table 5.2. The best result with the proposed method was a NRMSE of 0.0131 while with the TDNN 3 obtained a NRMSE of 3 The number of PEs in the hidden layer of the TDNN is chosen as 10 by 20 Monte-Carlo simulations varying the size of the hidden layer.

PAGE 74

62 0.0205. On the other hand, a single linear ARX model produced much higher NRMSE, 0.0486, than others did. This result shows that the proposed multiple linear modeling scheme outperforms both the linear and the nonlinear global modeling paradigm. Table 5-2. Comparison of modeling performance for the controlled Lorenz system. Methodology NRMSE ARX (1) 4.8e-2 Multiple ARX (64) 1.3e-2 TDNN (5:10:1) 2.1e-2 Next, we tested the proposed control scheme using multiple models. First, we let the system converge to the origin, which is one of the equilibria of the Lorenz system, starting from the initial state by the Multiple Inverse Controllers (MIC); the controller is activated at 5 sec and the results are shown in Figure 5-4(a) where the closed loop system response is seen to converge to the origin fairly well. This is relatively easy to control since the affine system (5.1) of zero dynamics is asymptotically stable at the equilibrium point. Also the figure shows the behavior of the control input. Second, we forced the controlled Lorenz system to a set point, Txxx]10,10,10[],,[321 8 dy which is not one of the equilibria. This is a more complicated task than steering the state of the system to the origin. The regulation results with the proposed MIC are shown in Figure 5-4(b) where we observe that the states and asymptotically regulate to and, respectively, and the state remains bounded. They illustrate that the conditions and described in [108] are satisfied even in the case under consideration. 1x 2x 81x 82x 3x 12xx /)(213xx

PAGE 75

63 (a) (b) Figure 5-4. Tracking a fixed point reference signal by MIC: (a) (b) 0dy 8dy Afterward, the MIC was investigated as increasing the number of controller and compared with a single linear inverse controller built by a ARX model as well as a nonlinear TDNN controller, which is a global controller trained through the TDNN model, for the same task. The optimal number of PEs in the hidden layer of the TDNN controller was selected as 40. Figure 5-5 illustrates how the number of controllers effects control performance. As expected, a single inverse controller based on a ARX model showed the worst performance with regard to settling time as well as steady-state error even though it reached the set-point the first time. As seen in the figure, the faster the rising time, the larger the overshoot, which is an unwanted factor in most cases, is

PAGE 76

64 generated. A smaller overshoot, faster settling time, and less steady-state errors were obtained by using a larger number of inverse controllers, such as the 16-MIC, which showed a very similar performance to the TDNN controller. Moreover, the response using the 64-MIC demonstrated very small steady-state errors. The 144-MIC, however, reached the set-point much slower than the others. It also took much longer to settle to the desired point. Too many divisions of state space may cause poor control performance (frequent switching among controllers) so that the controller may not be capable of following fast-changing trajectory. Figure 5-5. Comparison of control performance varying the number of inverse controllers based on multiple models. Finally, the proposed multiple control schemes, especially MIC, were compared with a linear controller and a nonlinear controller. The PID controller coefficients were determined to bring the poles of the closed-loop response from the plant output to the desired output to 0, 0, 0.55+i0.31, and 0.55-i0.31. The QSM controller was designed by

PAGE 77

65 choosing the switching surface with Tc]3,1[ and 001.0,8.0 rTqT Depending on the selection of the coefficients for PID and QSM controllers, one obtains the performance desired. This issue will be discussed in later sections. The step performance of the designed multiple model based control scheme and the TDNN controller in closed-loop operation with the controlled Lorenz system is provided for the two case studies, and 0dy 8 dy in Figure 5-6. Figure 5-6. Comparison of tracking performance by multiple model based controllers (MIC, MPIDC, MQSMC) and global inverse controllers (IC-ARX, TDNNC). We can observe from these results that the overshoots of the Inverse Controller based on the ARX model (IC-ARX) are much larger than others, which also results in a much longer settling time. Moreover, even if IC-ARX and TDNNC demonstrated very short reaching time, they exhibited relatively larger steady-state errors than the others. On the contrary, the multiple inverse controller showed a much faster reaching time, which is close to that by IC-ARX and TDNNC, with small steady-state errors when compared with MPIDC and MQSMC. These comparisons of IC-ARX, TDNNC, and MIC seem reasonable since they all have an inverse control framework. From these results, it can be easily inferred that the proposed control strategy guarantees the convergence of the

PAGE 78

66 system gaining a fast response to the desired set-point, even though the plant with the highly nonlinear characteristics is not known a priori to the controller. 5.1.2 The Duffing Oscillator Another system considered corresponds with the control of a Duffing oscillator, which displays chaotic behavior, described by tpxpxpxpuxxxcos43131221221 (5.3) where is a constant frequency parameter, ,, and are constant parameters [10,32,34,35]. 1p 2p 3p 4p Figure 5-7. The uncontrolled Duffing oscillator: phase-space trajectory and time-series. We assume that the controlled Duffing oscillator is originally ( 0 u ) in the chaotic state, shown in Figure 5-7, with parameters 8.1 [32]. In the simulations, the Duffing system was considered as unknown, which only generated time-series data excited by uniformly distributed control input, Tpppp]8.1,0.1,1.1,4.0[],,,[4321 5u via the fourth-order Runge-Kutta scheme with a fixed time step of 0.2. The embedding dimension for model construction was chosen as 1 ud and 1 yd based on the Lipshitz index in Figure 5-8. Thus the model is designed as ),,,(111 kkkkkuuyyfy Another

PAGE 79

67 parameter to be selected is the size of the SOM which is trained with Tkkykyy],[1 We chose the number of PEs in the SOM as 8 by the performance varying the size of the map shown in Figure 5-8. Figure 5-8. Lipschitz index (left) for the determination of optimal number of inputs and outputs and Generalization error v.s. Number of PEs (right). Table 5-3. Comparison of modeling performance for the controlled Duffing oscillator. Methodology NRMSE ARX (1) 3.4e-2 Multiple ARX (64) 1.3e-2 TDNN (4:14:1) 1.1e-2 The identification result with 8 square map and its performance comparison with one of global model, TDNN 4 is presented in Figure 5-9. In addition, another comparison with a single linear model is listed Table 5-3. We observe that the proposed multiple modeling strategy is a little worse than a TDNN, even if it demonstrated much better modeling performance than a single ARX model. The proposed modeling method, 4 A TDNN with 14 PEs in the hidden layer demonstrated best performance in modeling the Duffing oscillator.

PAGE 80

68 however, has a much simpler structure for modeling chaotic systems than a global modeling paradigm. Figure 5-9. Identification of the controlled Duffing oscillator by TDNN (left) and multiple-models (right). Next, we performed simulations for 3 different control tasks. 1) Control of a Duffing oscillator: For this task, we target the oscillator to follow an arbitrary trajectory generated by a random control input bounded by 5. 2) Synchronization of two Duffing oscillators: This is to show the proposed control scheme is able to synchronize oscillators effectively in spite of model mismatch; the parameters for the slave oscillator are 4.01 p 8.14 p and 8.1 whereas the parameters for the master oscillator are 41.01 p 24 p and 9.1 3) Synchronization of two strictly different second order oscillators: In this case, the slave oscillator is a Duffing one. For this oscillator, the parameters were taken as , and 4.01p 8.14p 8.1 The master oscillator is a van der Pol one which is described by 21xx )667.0(cos*75.1)1(1.0312212txxxx For these missions, the controllers were designed as follows: The optimal number of PEs in the hidden layer of the TDNNC was selected as 30 by 20 Monte-Carlo

PAGE 81

69 simulations. The MQSMC were built by choosing the sliding hyperplane with Tc]15,1[ and 001.0,8.0 rTqT The MPIDC were designed for placing the poles of the closed-loop response at 25.025.0i which demonstrated the fastest convergence to the desired trajectory in Figure 5-10. (a) (b) (c) (d) Figure 5-10. Performance comparison on trajectory tracking by TDNNC, PID-ARX, and MPIDC when the poles of the closed-loop response are place at (a) 0.9 (b) 0.50.5i (c) 0.25 0.25i (d) 0.05 0.05i. From these figures, it should be noted that the MPIDC produces shorter settling time than a single PIDC does regardless where the poles are placed. In addition, as the poles are getting closer to the unit circle or to the origin the convergence time is getting longer. Hence, the poles of the error dynamics should be chosen carefully.

PAGE 82

70 We compared the performance of the controllers using 50sec-long oscillatory trajectory regarding settling time and NRMS of steady-state error (NRMS-SSE) in Table 5-4: the settling time was selected when the tracking error was bounded in .1 and the NRMS-SSE was calculated using the tracking error from 30sec to 50sec. As seen in the table, the proposed control strategies outperformed a global controller, which is generally utilized for unknown system control, in both fast response and accuracy. Especially, while the TDNNC demonstrated some difficulty in following the desired path generated from a different dynamics, the van der Pol oscillator, the multiple controllers accomplished the mission relatively well. Table 5-4. Comparison of tracking performance for 3 different control task : Settling time and NRMS-SSE. Methodology Task 1 Task 2 Task 3 6.0sec 3.6sec 10.6sec TDNNC 2.8e-2 3.2e-2 6.8e-2 5.0sec 3.6sec 7.8sec MIC 2.3e-2 2.2e-2 2.7e-2 3.6sec 1.6sec 6.2sec MPIDC 1.9e-2 1.9e-2 1.8e-2 4.2sec 2.4sec 5.2sec MQSMC 2.7e-2 2.6e-2 3.2e-2 Moreover, the MPIDC among the proposed multiple control methodologies showed a significant reduction of the transient time as well as accuracy improvement in most missions. Figure 5-11 shows the synchronization of the master and slave oscillator signals by the TDNNC and MPIDC where the controller is activated at It is seen from the results again that the transient time is shortened without resulting in overshoots using the MPIDC. sec8t

PAGE 83

71 (a) (b) (c) Figure 5-11. Control performance by TDNNC (left) and MPIDC (right): The dotted line is the desired trajectory and the solid line is the output of the oscillator. (a) Tracking an oscillatory reference signal (b) Synchronization of two Duffing oscillators (c) Synchronization of a Duffing oscillator and a van der Pol oscillator.

PAGE 84

72 5.2 Nonlinear Discrete-Time Systems We have seen how effectively the PID and the inverse controller can be implemented using multiple models. In this section, nonlinear discrete time systems in a noisy environment are considered focusing more on multiple model-based control with sliding mode. 5.2.1 A First-order Plant Consider the following nonlinear discrete-time plant [5] kkkkkkkkkkzxyuxxxxxx,2,22,2,11,2,21,1)(1163 (5.4) where is the input and is an external disturbance. In (5.4), we considered a SISO model, assuming that only the output is available for measurement. The output time-series was created by exciting an input signal that is uniformly distributed in ku kz ky 5.0u (a) (b) Figure 5-12. Parameter selection to design multiple models: (a) Lipschitz index for determining the embedding dimension (b) Identification performance v.s. network dimension on independently generated test data for choosing the size of a map.

PAGE 85

73 The model was assumed to be a second order in input and output based on Figure 5-12(a). After quantization of the embedded output space y a set of models was built with the input-output data samples for each PE. For testing, 400 independently generated data samples were used changing the size of the map. The best size of the map was determined as 8 since the performance did not improve much after 64 PEs (see Figure 5-12(b)). Thus plant identification with 64 multiple models (8) was tested in the absence of sensor noise as well as in the presence of sensor noise with the plant input signal being uniformly distributed. The result of system identification in the absence of sensor noise by multiple models is shown in Figure 5-13(a). As we can see, the models provide a very good approximation of the plant visually based on the error signals. (a) (b) Figure 5-13. Modeling performance using 64 multiple models for a nonlinear first order plant: (a) System identification of the nonlinear plant in the absence of disturbance by the proposed multiple models. (b) Comparison of robustness against noise between TDNN and multiple models. Also, the proposed multiple modeling scheme was compared with a conventional TDNN that was trained through the backpropagation algorithm with 5000 samples and a constant learning rate, 0.005. The modeling result with 64 multiple models was a NRMSE

PAGE 86

74 of around 6.5e-4 while with the TDNN 5 one obtained a NRMSE of about 6.3e-3. The performance when the testing data are perturbed by noise is shown in Figure 5-13(b) where we observe that the SOM is more likely to select the wrong model as the noise level is increased. Even though the multiple models presented more accuracy in identification than a global model in the absence of noise, it should be pointed out that the multiple modeling strategy does not have more noise-immunity than global models at certain noise level. However, the proposed method is more robust than a global model up to certain noise level. Figure 5-14. Responses for parameter selection to design QSMC by varying (a) r T and (b) qT. Based on the 64 multiple models and the TDNN model identified, we designed multiple controllers and a TDNN controller, respectively. The TDNN controller was trained by back-propagating an error through the TDNN model taught by 20 hidden PEs. The number of hidden PEs in the controller was chosen as 40. The MPIDC was designed for placing the poles of the closed-loop response at 3.01.0i The MQSMC was built by choosing the sliding hyperplane with Tc]2,1[ and 01.0,8.0 rTqT comparing the 5 The number of PEs in the hidden layer of the TDNN is chosen as 20 by 20 Monte-Carlo simulations varying the size of the hidden layer.

PAGE 87

75 parameters as in Figure 5-14. In the figures, we observe that the larger the value of r T is the faster the plant reaches to the reference signal. In addition, the smaller the value of is the smaller the sliding mode band is. qT First, the performance of the controllers was tested on square-wave [1,-1.5, 1, 0] tracking in the absence of sensor noise and the results are presented in Figure 5-15 where we can see that all controllers showed very good performance on tracking the reference signal. Specifically, the MPIDC is the most accurate controller in spite of very slow convergence, and the MIC is the fastest one even though it shows a little steady-state error. In contrast, the MQSMC demonstrated the worst performance among the 4 controllers regarding rising time and steady-state error, but it shows its superiority in the presence of noise later. Figure 5-15. Comparison of tracking performance using a global controller and multiple model based controllers in the absence of sensor noise. The figure (right) is an enlargement of the figure (left) between 34 and 52 iterations. Figure 5-16 shows the plant responses of the closed loop control system using the MQSMC. The trajectories are seen to converge to the desired values of [-1.5 -1.0 -0.5 0.5 1.0 1.5]. The figure also shows control input, the sliding surface, and the winner activities switched automatically by the SOM. It can be easily seen that the proposed MQSMC

PAGE 88

76 scheme guarantees the convergence of the system to the quasi-sliding-mode band around the sliding hyperplane 0kTec Figure 5-16. Performance of square-wave tracking in the absence of noise by the MQSMC. Next, the robustness of the proposed control scheme was compared with that of a global controller using TDNN. The standard deviation of the error between the plant output and the desired output versus the standard deviation of the noise is shown in Figure 5-17. It is evident that the MQSMC performs best in terms of insensitivity to disturbances. The MIC structure showed the best performance only in the noise-free environment. It should be noted that the MIC and the MPIDC began to become less robust than the TDNNC at the point where the standard deviation of the noise is over 0.04. From this examination we can conclude that the MIC and the MPIDC can be robust

PAGE 89

77 against noise up to certain level. However, wrong selection of the winner, as the amount of noise is increased, can be devastating for the controller that is designed based on the predicted model. 00.020.040.060.080.10.120.140.160.180.0000.0030.0110.0230.0460.0700.0930.115Standard deviation of noiseStandard deviation of tracking erro r TDNNC MIC MPIDC MQSMC Figure 5-17. Comparison of performance against noise among TDNNC, MIC, MPIDC and MQSMC. Furthermore we tested the closed loop system for tracking a sinusoidal and an arbitrary desired output. Once again, the multiple controller networks perfectly track the desired command except for a transient time of a few time steps, as shown in Figure 5-18, even if the measurement is corrupted by zero-mean random noise with 20dB of SNR. Overall, we conclude that the proposed MQSMC approach is the most robust design technique among the four methods considered. This is evident from Figure 5-17, where we observe that on average the tracking error of the MQSMC increases at a lower rate than that of the MIC, the MPIDC, and the TDNNC.

PAGE 90

78 (a) (b) Figure 5-18. Sinusoidal and arbitrary signal tracking by the MQSMC:(a) in the absence of sensor noise (b) in the presence of sensor noise, dBSNR20 5.2.2 A Laboratory-scale Liquid-level Plant A liquid-level system described by the following second-order equation has been considered: 111112121121111087.03084.003513.003259.01663.004228.03103.01295.03578.09722.0kkkkkkkkkkkkkkkkkkkuuyuyyuyyyuyyuyuuyy (5.5) This model is obtained through identification of a laboratory-scale liquid-level system [80,1]. In [80], this model has been used to illustrate theoretical developments for direct adaptive control. 10,000 data samples were generated for analysis using the control

PAGE 91

79 effort, 1ku which has uniform distribution. Given the prior information concerning the order of the plant by Lipschitz index in Table 5-5, a second order input-output model described by the following equation was chosen to identify the plant: ),,,(111 kkkkkuuyyfy (5.6) Table 5-5. Lipschitz index of a laboratory-scale liquid-level plant for determining an embedding dimension. Number of inputs 1 2 3 4 5 1 262.34 10.86 1.80 1.64 1.46 2 13.49 1.32 1.18 1.14 1.13 3 1.46 1.16 1.13 1.13 1.12 4 1.19 1.15 1.10 1.10 1.09 Number of outputs 5 1.19 1.15 1.09 1.09 1.09 (a) (b) Figure 5-19. Modeling performance using multiple models for a liquid-level plant: (a) Identification performance v.s. network dimension for choosing the size of a map of the liquid-level plant (b) Identification of a liquid-level plant using 12 multiple models. The embedded output vector, ky, was used for SOM training over 6,000 epochs, and then not only ky, but the embedded input vector, ku, were exploited to create local linear models. In order to test the performance of the proposed modeling scheme,

PAGE 92

80 we applied 400 newly generated data samples to multiple models. Figure 5-19 depicts identification performance depending on the size of the network and how well the multiple models approximate the liquid-level plant using 144 models. In addition, we compared the identification performance using multiple models with that by a TDNN. The number of inputs and outputs to the network were the same as in local linear modeling. A single hidden layer of 30 PEs was large enough for good identification performance by 20 Monte-Carlo simulations varying the size of the hidden layer. As seen in Table 5-6, the proposed multiple models slightly outperformed a nonlinear model in a liquid-level plant identification. Also the table clearly shows the benefit of using multiple models when comparing with the performance by a single ARX model. Table 5-6. Comparison of modeling performance for the liquid-level plant. Methodology NRMSE ARX (1) 29.2e-3 Multiple Models (12) 3.5e-3 TDNN (4:30:1) 4.3e-3 A typical open-loop input-output characteristic of the plant is shown in Figure 5-20. The large variations in the steady-state gain and time constant with the operating point is also clearly visible from this figure. The performance of the proposed multiple model based controllers were compared to that of a nonlinear TDNN controller designed through the previously identified TDNN model. The controllers for this plant were designed as follows: First, the optimal number of PEs in the hidden layer of the TDNN controller was chosen as 50. Second, the PID controller was designed in order for the closed loop response poles to be located in 0.7702 and 0.1298. Third, the sliding

PAGE 93

81 hyperplane of the QSM controller was set to Tc]66.1,1[ based on Figure 5-21 which illustrates the effect of the parameter selection for the sliding surface design under different noise levels. Figure 5-20. Typical input-output characteristic of the second-order liquid-level plant. Figure 5-21. Square-wave tracking performance of the liquid-level plant varying the sliding surface and the noise level by the MQSMC.

PAGE 94

82 In most cases, as we placed the poles ( 21/cc ) closer to the origin inside the unit circle, the controller showed better tracking performance. For instance, from the plot, we can say that the pole should be chosen as between 0.5 and 0.6 to have the robustness against noise whose level is 25dB of SNR since the error changes too slowly above 0.8 and very fast below 0.5. Thus the switching surface was chosen as in order to get small error and short enough transient time as well. Tkkkees],][66.1,1[1 (a) (b) (c) (d) Figure 5-22. Control of the liquid-level plant by the MQSMC varying the number of controllers: (a) M = 1 (b) M = 16 (c) M = 36 (d) M = 144. The control systems were tested for a typical square wave set-point of amplitude 0.5. Figure 5-22 compares the tracking performance of the reference signal, which is the output of a first-order reference model with transfer function and )8.0/(2.0)(zzzGd

PAGE 95

83 driven by the set-point, by the MQSMC varying the size of the map. The figure illustrates how the control performance is affected by the number of QSM controllers. As the number of controllers is increased the plant reaches the set-point with smaller steady-state errors. Even if there seems, however, no big difference between Figure 5-22(c) and Figure 5-22(d) the larger number of controllers might be working better in a noisy environment. This will be discussed again later in this section. Additionally, a comparison of this figure with Figure 5-20 displays the open-loop response and establishes the efficacy of the proposed control scheme. Table 5-7. Comparison of control performance for the liquid-level plant in noise-free environment. Methodology NRMSE TDNNC (4:50:1) 5.7e-3 MIC (12) 1.4e-3 MPIDC (12) 10.9e-3 MQSMC (12) 5.0e-3 Table 5-8. Comparison of control performance for the liquid-level plant in the presence of sensor noise: standard deviation of noise is 4.5e-2. Methodology Standard deviation of error TDNNC (4:50:1) 3.8e-2 MIC (12) 4.2e-2 MPIDC (12) 10.2e-2 MQSMC (12) 1.9e-2 Table 5-7 compares the tracking performance among controllers where MIC and MQSMC outperformed TDNNC. However, even though MPIDC showed the worst performance it showed the smallest steady-state errors among them. It only demonstrated

PAGE 96

84 large errors in the transient phase since it was not fast enough to track the trajectory. In addition, the controllers were tested on the same tracking trajectory regulating disturbances whose standard deviation is 4.5e-2. The results are listed in Table 5-8. (a) (b) (c) (d) Figure 5-23. Control of the liquid level system with measurement noise by the MQSMC with (a) M = 1 (b) M = 16 (c) M = 144 and (d) the TDNNC. As in the previous result for controlling the first order plant, the MQSMC showed the best performance in robustness against the measurement noise, and the MPIDC showed the worst performance again due to the slow convergence. Figure 5-23 compares the response among a nonlinear controller and the MQSMCs varying the number of the controllers under a noisy environment. As we can see, the proposed MQSMC is capable of rejecting measurement noise and following the trajectory with very little variations.

PAGE 97

85 Also the MQSMC with 144 M showed a more stable aspect than the smaller structure. While a nonlinear controller, the TDNNC, showed relatively large chatters on the trajectory even if it tracks the path, it also spent much more control effort to do the same mission than the MQSMC did. Figure 5-24. Tracking an oscillatory reference signal of the liquid-level plant by the TDNNC (left) and the MQSMC (right) in the presence of sensor noise. Figure 5-25. Performance assessment on a trajectory tracking under noisy environment. Finally, we evaluated the tracking performance of an oscillatory reference signal by the TDNNC and the MQSMC depicted in Figure 5-24. The reference signal to be followed was ))1.0cos(2.1)2/15.0sin(8.0(2.0kk We observed that the MQSMC tracked the trajectory more smoothly than the TDNNC did. In summary, the proposed MIC Measurement noise O p en-loo p Control Perfect Cotnrol M Q SMC TDNNC MPIDC 0 Increasing closed-loop variance

PAGE 98

86 MQSMC is the best possible multiple model based controller in a noisy environment for both trajectory tracking and set-point tracking as illustrated in Figure 5-25. On the other hand, the MPIDC is not suitable for a trajectory tracking problem that requires fast transient response coping with measurement noise even if it is the best in set-point tracking in the absence of noise. 5.3 Flight Vehicles Flight vehicles, such as missiles and aircraft, are very complex systems that are typically non-minimum phase and have aerodynamic coefficients which vary over a wide dynamic range due to large Mach-altitude fluctuations [43,47]. Control of high-performance low-cost UAV especially involves the problems of incomplete measurements, external disturbances and modeling uncertainties. Nevertheless, the autopilot for these vehicles is often required to achieve very stringent performance objectives. Thus the proposed control algorithms are applied to missiles and UAV to show the effectiveness in this section. 5.3.1 Missile Dynamics We consider a benchmark missile model used widely in the literature [50,104]. This benchmark model can be formulated as 432101ggqggq (5.7) and 24322221542.14)cos(0403.0)]3/87(6001.33765.10152.0[)cos()]3/2(2010.00112.00001.0[MgMgMMgMMg

PAGE 99

87 where is the angle of attack (deg), is the pitch rate (deg/s), q M is the Mach number, and is the tail deflection angle (deg). We considered an operation range of 3 M and ]8,8[ deg. The model was formulated as 111,,, kkkkkuuyyfy where is the Angle Of Attack (AOA) to be controlled. ky (a) (b) Figure 5-26. Modeling performance using multiple models for the missile dynamics: (a) Identification performance varying the network dimension (b) Identification of a missile dynamics by 18 multiple models. The SOM was trained with the vector Tkkyy1, varying the size of the map and tested with 400 newly generated data samples. Identification of the missile system with 18 multiple models, which was chosen as the optimal dimension of network as shown in Figure 5-26(a), is depicted in Figure 5-26(b) where we observe that the models present a very good approximation of the plant. Table 5-9. Comparison of modeling performance for the missile system. Methodology NRMSE Multiple Models (18) 6.1e-3 TDNN (4:35:1) 6.3e-3

PAGE 100

88 The modeling performance using the designed multiple models were compared with that by a TDNN, which was trained by one hidden layer with 35 PEs, in Table 5-9. The proposed method showed marginally better performance than the nonlinear model did. Based on 324 linear models and a nonlinear TDNN model with 35 PEs in the hidden layer, we designed the multiple model-based controllers and the TDNN model based controller. The optimal parameters selected for each controller were 30 hidden PEs for TDNN controller, the desired pole locations of 0.3 and 0.7 for the closed loop response by the PID controller, and 11 c 85.12 c 51.0 qT for QSMC. (a) (b) (c) (d) Figure 5-27. Tracking various set-point reference signal by (a) TDNNC (b) MIC (c) MPIDC and (d) MQSMC in the absence of measurement noise.

PAGE 101

89 The performance of the proposed control scheme in the absence of noise was evaluated by comparing it with that of the TDNN controller for the tracking of AOA in Figure 5-27, in which the MPIDC showed the smallest steady-state error although its response is sluggish and the MQSMC produced the shortest settling time. Both of them did not have any overshoot. On the other hand, the TDNNC and the MIC, which have similar responses, generated unnecessary overshoot; consequently, it took longer than other controllers to settle down. As a result, the proposed MQSMC is capable of achieving precise tracking objective with a sufficiently fast response characterized by c (a) (b) (c) (d) Figure 5-28. Tracking various set-point reference signal by (a) TDNNC (b) MIC (c) MPIDC and (d) MQSMC under the presence of measurement noise whose standard deviation is 4.6e-2.

PAGE 102

90 In practical implementations of control structures for trajectory control, one difficulty in achieving accurate trajectory tracking is the existence of observation noise. To study the effects of this situation, which is very likely to be encountered in practice, the information used by the controller is corrupted by a random noise having zero mean and variance equal to 2e-3 and 5.3e-2 for the worst case. The simulation results in Figure 5-28 show that, with appropriate choices of the sliding surface and parameters, robust tracking can be obtained by the MQSMC even in the presence of sensor noise. Table 5-10. Comparison of controller performance for the missile system in the presence of noise having the standard deviation of 4.6e-2. Controller Standard Deviation of error TDNNC 10.44e-2 MIC 18.35e-2 MPIDC 10.99e-2 MQSMC 4.40e-2 Figure 5-29. Trajectory tracking by TDNNC (left) and MQSMC (right) in the presence of noise whose standard deviation is 2.3e-1. Additionally, we compared the controller performance for the task of set-point tracking of AOA starting from 0 degree to 4 degree in the presence of noise having the

PAGE 103

91 standard deviation of 4.6e-2 and listed in Table 5-10, in which we observe that the MQSMC is the only controller which produced less standard deviation of error than the standard deviation of noise. Figure 5-29 compares the trajectory tracking behavior of the MQSMC with the TDNNC, in which the MQSMC demonstrates the superior performance for disturbance rejection compared to its counterpart. Figure 5-30. Set-point tracking behavior by the TDNNC (left) and the MQSMC (right) under parameter variations. Finally, we regulated AOA to stay 0 degree varying Mach number from 1 to 3 in order to test the controller performance under parameter variations. From Figure 5-30 it is clear that the TDNNC suffers from chattering, especially, at the point where the Mach number is changed, it showed high overshoot. In contrast, the MQSMC demonstrates its excellent robustness despite parameter variations without any vibration due to its ability of staying in the switching surface and prompt switching of controllers. Consequently, the proposed MQSMC provides a robust control system in the presence of either an external disturbance or parameter variations.

PAGE 104

92 5.3.2 LoFLYTE UAV The second example of the flight vehicle considered for this study is the LoFLYTE UAV designed by Accurate Automation Corporation (AAC), shown in Figure 5-31. Figure 5-31. General description of aircraft (left) and LoFLYTE testbed UAV (right). The LoFLYTE program is an active flight test program at the Air Force Flight Test Center at Edwards Air Force Base, with the objective of developing the technologies necessary to design, fabricate, and flight test a Mach 5 waverider aircraft [18,19]. In addition, the LoFLYTE UAV has been used to understand the low speed characteristics of a hypersonic shape and to demonstrate several innovative flight control technologies. In classical notation, longitudinal motion consists of pitching ( q, ) motion, while the lateral motion consists of rolling ( p, ) and yawing ( r ) movement. The elevator ( e ) and the throttle ( th ) control the longitudinal motion, while the aileron ( a ) and rudder ( r ) primarily affect lateral motion. The general dynamics of the system are described by

PAGE 105

93 seccossecsinsincostancostansin/)()(/)()(/)()(/coscos)(/sincos)(/sin)(22rqrqrqpINqrpIpqIIrIMprIrpIIqILpqrIqrIIpMFguqvpwMFgwpurvMFgvrwquzzxzyyxxyyxzxxzzxxxzzzyyazayax (5.8) where is the aircraft mass and are actuator-induced forces and moments. The forces and moments are nonlinear functions of an aircrafts states and control inputs [98]. This model was programmed in a software simulator by AAC and is used here as the source of the data. In this study, we wish to estimate and control the aircrafts lateral motion under the assumption that we can only access roll-rate (p) and yaw-rate (r), while the goal is to track the desired trajectories ( and ) during the course of the flight considering the case of an aircraft moving with a constant throttle. aM NMLFFFzyx,,,,, dp dr To model the aircraft lateral dynamics, a total of 2 SOMs are used for quantization of each embedded output as predictors, Tdykkkyyy1,,, because as the number of inputs to the SOM is increased, the SOM becomes increasingly inaccurate to quantize in the feature space. Thus the linear coupling between w v u r q p ,,,,, is only implicitly modeled 6 In this way, each output (either p or r ) of the aircraft can be described by a dynamic model that takes into account the control input variables such as ra ,: 6 As we mentioned in Chapter 2, due to difficulties related with dynamic range normalization, local linear models that take state coupling into account are not as accurate as this approach. Instead, we utilize the delayed outputs in order to compensate for the disregarded information due to the coupling.

PAGE 106

94 Nifykykuik,,1),,(,,1 (5.9) where Tdukrkrdukakaku1,,1,,,,,,,, By doing this, the complexity can be reduced and it helps to understand the raw data. Again, we selected an embedding dimension based on the Lipschitz index. Since the optimal values of embedding dimension for each output dynamic model are different, we chose the largest number among the embedding dimensions selected for each output estimation as for each output and 2dy 6 du for 2-D control inputs (aileron and rudder). The linearized input/output relationship then is 611,,21,,12111611,,21,,12111iikrriikariiikrikiikrpiikapiiikpikbbrarbbpap (5.10) Figure 5-32. Control inputs ( ra ) used to generate data samples for training the networks. The training samples were obtained by exciting the aircraft dynamics (LoFLYTE Simulator) running at a frequency of 10 Hz using the control inputs shown in Figure 5-32. Each SOM is trained with the embedded output, y whose dimension is 2, over 5000 samples and the reasonable result for identification of the roll-rate was obtained with a

PAGE 107

95 66 grid map () as shown in Figure 5-33 where also the modeling performance with the selected number of models is depicted. 36N (a) (b) Figure 5-33. Modeling of a roll-rate using multiple models: (a) Identification performance varying the network dimension (b) Identification of a roll-rate ( p ) dynamics by 6 multiple models. Table 5-11. Comparison of modeling performance for the lateral motion ( p and r ) of the LoFLYTE UAV. NRMSE Methodology roll-rate ( p ) yaw-rate ( r ) Multiple models (6) 4.7e-3 3.8e-3 TDNN (16:50:2) 7.9e-3 3.9e-3 After training the SOMs, 36 local linear models were constructed and the created models were tested by new sequence with 1,000 samples. Table 5-11 shows the identification performance of two dynamics of the system with 36 models and compares their performances with that from a TDNN model. Training conditions for a TDNN model, such as the embedding dimension was kept the same in this comparison between the local modeling and the global modeling. The best result with a TDNN model was

PAGE 108

96 obtained from 20 Monte-Carlo simulations with 50 PEs in the hidden layer 7 From the table, we can conclude that the constructed SOM-based network is a good model of the underlying dynamics because it provides smaller NRMSE for all dynamics than the TDNN model. Consequently, it turned out that the proposed strategy of finding proper location of fixed models depending on the prior information available to the designer for finding aircraft dynamics is superior to those using a single global nonlinear model. In addition, it should be noted that the proposed modeling scheme makes identification of the plant very compact and computationally efficient since the aircraft dynamics are captured in a compact lookup table of linear models. We now consider the control problem with the SOM-based local models created. When we design controllers we usually assume that the coupling between lateral and longitudinal motion is minimal. Here, we performed a simulation to control the roll-rate ( p ) and yaw-rate ( r ) of the aircraft by aileron ( a ) and rudder ( r ), setting elevator to zero and throttle to constant. Thus, once we have the linear models for the roll-rate and the yaw-rate, and the desired values, and the inverse controller (inversion-based predictive model), 1,kdp 1,kdr ka, and kr, for the aircrafts roll-rate and yaw-rate tracking is obtained by 5,1,5,1,11,5,1,5,1,11,1,,,,,,,,,,,,,,,,,,krkrkakakkkdkrkrkakakkkdikrkarrrpppf (5.10) And the sliding mode controller was designed such that the poles of the error dynamics are placed at 0.5 and other parameters are set to 5.0 qT and Also, for 001.0rT 7 We tried various sizes of TDNNs and found that the one with 12 PEs in the hidden layer performs best in system identification.

PAGE 109

97 performance comparisons, we applied the TDNN controller, which has 100 PEs in the hidden layer, for the same control problems. Figure 5-34 and Figure 5-35 compare the set point tracking performance of the TDNNC, the MIC, and the MQSMC in the absence of sensor noise and in the presence of noise whose level is 20 dB of SNR, respectively. From the responses it can be seen that the multiple controller approach is very good except for the first few seconds. However, it shows poor transient response when the global control, the TDNNC, is utilized. Another performance test is to enforce the tracking of the roll-rate and yaw-rate to signals and which are given in real time during the course of the flight, while being subjected to unmeasured sensor disturbances. The 2 output measurements are corrupted by zero-mean random sequences with 20 dB of SNR. The results of a flight test with the proposed method are shown in Figure 5-36 where we also show the same with the TDNNC. It can be seen that the roll-rate and the yaw-rate track their command signals quite well even under the existence of measurement noise by the multiple controllers. The simulated flight test demonstrates that the proposed controller is capable of closely approximating the given mission by only looking at the past information. Also, it proves that the multiple controller framework indeed provides exceptional tracking. 1,kdp 1,kdr

PAGE 110

98 (a) (b) (c) Figure 5-34. Comparison for controlling roll-rate and yaw-rate to track the set point in the absence of noise by (a) TDNNC (b) MIC (c) MQSMC.

PAGE 111

99 (a) (b) (c) Figure 5-35. Comparison for controlling roll-rate and yaw-rate to track the set point in the presence of noise by (a) TDNNC (b) MIC (c) MQSMC.

PAGE 112

100 (a) (b) (c) Figure 5-36. Performance of controlling roll-rate and yaw-rate to track an arbitrary trajectory with measurement noise (SNR = 20) by (a) TDNNC (b) MIC (c) MQSMC.

PAGE 113

CHAPTER 6 CONCLUSIONS AND FUTURE WORK In this dissertation, a novel approach to the modeling and control of complex nonlinear plants was developed. The modeling technique is local and empirical in the sense that only input-output data from the experiments were used to identify the model, and the structure of the model is controller orientated. It was shown how the design of multiple controllers can be based on the model structure presented. 6.1 Summary The problem of nonlinear system identification and control system design was addressed under the divide-and-conquer principle. Classical control approaches involving linearization and gain scheduling in which the parameters of the controller are scheduled according to changes in system dynamics are seldom applicable to complex nonlinear systems since there are no general rules for designing gain scheduling regulators and a general approach. However, an obvious way to analyze complex systems is to adopt the divide-and-conquer type strategy. This principle motivated the use of multiple local models for system identification in order to simplify the modeling task. The operating region is decomposed into smaller sub-regions, which are then described by local models of simple, possibly linear structure. Especially in the case of unknown dynamics, where only input-output data from the plant are available, the proposed method is able to approximate the nonlinear dynamics of the plant using a piece-wise linear dynamical model that is optimized solely from the available data. In particular, when local linear models are used as described, it also 101

PAGE 114

102 became possible to design a piece-wise linear controller for the plant, whose design is based on the identified model. The concept of self-organization was taken in embedded output space extended with multiple models by the SOM for system identification. The significance of the proposed scheme is that the operating region is selected by the embedded output and that local models are built by the embedded output as well as the embedded control input data samples which are spaced in the local area, which marks the fundamental difference between this work and others. The effectiveness of the proposed approach was shown through experiments for modeling complex nonlinear plants such as chaotic systems, nonlinear discrete time systems and flight vehicles. Its comparison with neural networks-based alternatives (e.g., TDNN) showed clear advantages of local modeling in terms of performance. An added advantage of the proposed local linear modeling approach is it greatly simplifies the design of control systems for nonlinear plants. In general, this is a daunting task and typically practical solutions involve linearization of the dynamics and then employing well-established controller design techniques from linear control systems theory. While designing globally stable nonlinear controllers with satisfactory performance at every point in the state space of the closed loop control system is extremely difficult, and perhaps impossible to achieve especially in the case of unknown plant dynamics, by using the local linear modeling technique presented, coupled with strong controller design techniques from linear control theory and recent theoretical results on switching control systems, it becomes possible to achieve this goal through the use of this much simpler approach of local modeling.

PAGE 115

103 The multiple model based linear controllers, such as inverse controller and PID controller, were introduced for a general class of nonlinear unknown discrete-time systems via SOM, which divides the state space into a set of operating regions. In addition, the multiple quasi-sliding mode control strategy was proposed. Contrary to what is known in the field of sliding mode controller design, the plant dynamics under control are assumed to be unknown. This is a challenge in the conventional design framework with the ambiguities introduced by the noise on the measured quantities. The problems that arise due to the uncertainties of the plant model and measurement noise are alleviated by incorporating the robustness provided by the sliding mode technique into the multiple modeling approach. The simulation results demonstrated that the algorithm proposed is able to compensate deficiencies caused by the imperfect observations of the state variables and complex plant dynamics, driving the tracking error vector to the sliding manifold and keeping it on the manifold. In addition, the proposed method shows better robustness against noise, faster transient response, and better steady-state accuracy of the controlled system by switching local controllers astutely through the SOM than other neural network-based alternatives. 6.2 Future Work The research efforts described in this dissertation were aimed at the development of a local modeling and control paradigm, and it has been justified by experiments with various nonlinear plants. During the process of this research work, a number of issues have arisen to achieve a better modeling and control performance. The model-controller pair has been selected by looking only at the history of the plant output in order to avoid some difficulties such as normalization of the input-output joint space and large dimensionality of the space involved. The system dynamics, however, may be captured more effectively in the joint space. Thus,

PAGE 116

104 efficient partitioning the operating regions in high dimensional joint space would be helpful to obtain a more accurate model. There is no theoretical framework based on which appropriate values for the parameters in SOM training can be chosen, e.g., initial value for the learning rate and width of the neighborhood functions, and subsequent rate of decrease and shrinkage, respectively. Hence, how to more effectively quantize the state space by SOM remains for further study. An inverse framework has been employed to design multiple controllers using identified models. In spite of its simplicity and fast response, some assumptions should be accompanied: the model is very precise and noise or external disturbances are negligible. If there is disturbance, then the plant output tracks a signal which is equal to the desired output plus the disturbance. For this reason, adding an adaptive scheme or a disturbance canceller, which feeds back an estimate of the disturbance to the controller of the multiple inverse control strategy, is required for robust control performance. To make use of inverse control, the plant must be stable or it must be stabilized. Real-time stabilization of an unknown, unstable and possibly nonlinear plant is a difficult problem, and should be explored. It has been shown that a multiple sliding mode control scheme guarantees BIBO stability of the overall system. We need to further investigate the stability conditions for switched control systems. In sliding mode controller design using multiple models, the sliding hyperplane has been determined such that the roots of the closed loop response are inside the unit circle, which guarantees an asymptotic convergence to the desired output. The closed loop system, however, may have very fast transient response, and possibly too large an unexpected overshoot depending on the measurement noise level. Therefore, it would be valuable if an advanced structure for determining sliding surface considering disturbances is invented. 6.3 Concluding Remarks We have presented multiple model based control strategy for tracking control of general unknown nonlinear systems through approximating the state space using a set of local linear models. The simulation results showed that the derived control using multiple models turns out to be highly promising and easy to implement in control engineering practice. These are encouraging results that motivate the use of this modeling and control technique for various other control applications. The capabilities of the local linear

PAGE 117

105 modeling approach is not limited to system identification and control applications. There are a wide range of nonlinear signal processing problems, such as magnetic resonance imaging, speech processing, and computer vision, where the concept of local modeling can be employed to obtain simple but successful solutions to difficult nonlinear problems.

PAGE 118

APPENDIX QSMC FOR MIMO SYSTEM Consider one of the local l multi-input multi-output models of the plant if f 1,,,,1,,1,1,1,,11,1,1,,,1,1,1,,,1,1,,11,1,1,1,11,1,11,1,1,11,11,1 nklnllklllnknlklmklmlkllklnklnlkllnknkmkmkkububububyayayububububyayay (A.1) where available measurement and the embedding dimensions of output and input are and respectively. The state-space model of (A.1) can be written as lyu, m n 11211 nknkkxkuuuxx (A.2) and and have the following forms: n,,1 1,1,,1,11,1,110000101000010lmlmlmmaaaaaa00 nllnlnlnnllllllllbbbbbbbbbbbb,,,1,,,1,1,12,,2,1,2,,12,1,121,,1,1,1,,11,1,11,,,,,,,,,,,,000000 106

PAGE 119

107 The equivalent control produces the following equivalent system kTTkxCCIx ])([1111 (A.3) and the problem is to find such that the eigenvalues of the matrix are all inside the unit circle. TC ])([111TTCCI For simplicity, we set 2 l If the switching surface parameters are defined as mllmTccccC,1,,11,100 (A.4) then 1,,,1,1,,1,,1,11,1,1,11llmllmllmmTbcbcbcbcC To simplify the problem we choose )(/1,1,11,,11,1,1,,1,1,1,1,,11,,1,11,1,,1,,,1,1,1,1llllmmlmlmlmlllmlmbbbbcccbcbcbcbc This produces kTTkxCCIx ])([1111 subject to 0 ks in which case 0,2,1,,1,21,2,11,1kmkmkmkmmkTxxxxccccxC00 (A.5) kmkmkmkmmkmkmmmxxxxccccxxcc,12,1,1,11,21,21,11,1,2,,2,10000

PAGE 120

108 Solving for and yields kmx, kmx,2 kmkmkmkmmmmmmkmkmxxxxccccccccxx,12,1,1,1,21,2,21,2,11,1,11,1,2,00 (A.6) The reduced order equivalent linear system is kkmkmkmkmmmmmmkmkmkmkxxxxxccccccccxxxx,12,1,1,1,21,2,21,2,11,1,11,11,121,11,11,110001010001000 (A.7) The characteristic polynomial of the equivalent system is )det()(I Given the desired characteristic polynomial as we can compute by comparing 2212122)(mmm TC )( with )(

PAGE 121

LIST OF REFERENCES 1. Ahmed, M.S., Design of a Fixed Neural Controller for Nonlinear Plants, International Journal of Control, 69(1):65-83, 1998. 2. Bartoszewicz, A., Discrete-Time Quasi-Sliding-Mode Control Strategies, IEEE Trans. Industrial Electronics, 45(4):633-637, 1998. 3. Bottou, L. and V. Vapnik, Local Learning Algorithms, Neural Computation, 4(6):888-900, 1992. 4. Brown, R.E., G.N. Maliotis, and J.A. Gibby, PID Self-Tuning Controller for Aluminum Rolling Mill, IEEE Trans. Industry Applications, 29(3):578-583, 1993. 5. Camps, J., F.L. Lewis, and R. Selmic, Backlash Compensation with Filtered Prediction in Discrete Time Nonlinear Systems by Dynamic Inversion Using Neural Networks, Proceedings of the IEEE Conf. on Decision and Control, 4: 3354-3540, 2000. 6. Cao, L., Practical Method for Determining the Minimum Embedding Dimension of a Scalar Time Series, Physica D, 110(1):43-50, 1997. 7. Casdagli, M., Nonlinear Prediction of Chaotic Time Series, Physical Review D, 35(3):35-356, 1989. 8. Chakraborty, G. and B. Chakraborty, A Novel Normalization Technique for Unsupervised Learning in ANN, IEEE Trans. Neural Networks, 11(1):253-257, 2000. 9. Chan, C.Y., Robust Discrete Quasi-Sliding Mode Tracking Controller, Automatica, 31(10):1509-1511, 1995. 10. Chen, G., G. Chen, and R.J.P. Figueiredo, Feedback Control of Unknown Chaotic Dynamical Systems Based on Time-Series Data, IEEE Trans. Circuits and Systems-I, 46(5):640-644, 1999. 11. Chen, L. and K.S. Narendra, Intelligent Control Using Neural Networks and Multiple Models, Proceedings of the American Control Conference, 2:1357-1362, 2002. 109

PAGE 122

110 12. Chen, X., T. Fukuda, and K.D. Young, Adaptive Quasi-Sliding-Mode Tracking Control for Discrete Uncertain Input-Output Systems, IEEE Trans. Industrial Electronics, 48(1):216-224, 2001. 13. Cho, J., J. Lan, G. Thampi, J.C. Principe and M.A. Motter, Identification of Aircraft Dynamics Using a SOM and Local Linear Models, Proceedings of the IEEE Int. Midwest Symposium on Circuits and Systems, 2:148-151, 2002. 14. Cho, J., J. Lan, J.C. Principe, and M.A. Motter, Modeling and Control of Unknown Chaotic Systems via Multiple Models, To be appeared in Proceedings of the Int. Workshop on Machine Learning for Signal Processing, 2004. 15. Cho, J., J.C. Principe and M.A. Motter, A Local Linear Modeling Paradigm with a Modified Counterpropagation Network, Proceedings of the Int. Joint Conference on Neural Networks, 1:34-38, 2003. 16. Cho, J., J.C. Principe, and M.A. Motter, Local Hammerstein Modeling Based on Self-Organizing Map, Proceedings of the Int. Workshop on Neural Networks for Signal Processing, 1:20-24, 2003. 17. Cho, J., J.C. Principe, D. Erdogmus, and M.A. Motter, Modeling and Inverse Controller Design for an Unmanned Aerial Vehicle Based on the Self-Organizing Map, submitted to IEEE Trans. Neural Networks, 2003. 18. Cox, C., J. Neidhoefer, R. Saeks, and G. Lendaris, Neural Adaptive Control of LoFLYTE, Proceedings of the American Control Conference, 4:2913-2917, 2001. 19. Cox, C., R. Saeks, C. Lewis, W. Sefic, A. Carlton, M. Kocher, and R. Pap, Neural Control of the LoFLYTE Aircraft, Proceedings of the Int. Joint Conf. on Neural Networks, 2:1047-1050, 1998. 20. Diao, Y. and K.M. Passino, Intelligent Fault Tolerant Control Using Adaptive Schemes and Multiple Models, Proceedings of the American Control Conference, 4:2854-2859, 2001. 21. Efe, M.O. and O. Kaynak, Identification and Control of a Nonlinear Bioreactor Plant Using Classical and Dynamical Neural Networks, Proceedings of the IEEE Int. Symposium on Industrial Electronics, 3:1211-1215, 1997. 22. Eskinat, E., S. Johnson, and W.L. Luyben, Use of Hammerstein Models in Identification of Nonlinear Systems, AIChE Journal, 37:255-268, 1991. 23. Farmer, J.D and J.J. Sidorowich, Predicting Chaotic Time Series, Physical Review Letters, 59(8):845-848, 1987. 24. Foss, B.A., I.A. Johansen, and A.V. Sorensen, Nonlinear Predictive Control Using Local Models Applied to a Batch Fermentation Process, Control Engineering Practice, 3:389-396, 1995.

PAGE 123

111 25. Furuta, K., Sliding Mode Control of a Discrete Time Systems, Systems & Control Letters, 14:145-152, 1990. 26. Furuta, K., VSS Type Self-Tuning Control, IEEE Trans. Industrial Electronics, 40(1):37-43, 1993. 27. Gao, W. and J.C. Hung, Variable Structure Control of Nonlinear Systems: a New Approach, IEEE Trans. Industrial Electronics, 40(1):45-55, 1993. 28. Gao, W., Y. Wang, and A. Homaifa, Discrete-Time Variable Structure Control Systems, IEEE Trans. Industrial Electronics, 42(2):117-122, 1995. 29. Gawthrop, P.J. and E. Ronco, Local Model Networks and Self-Tuning Predictive Control, Technical report: CSC-96001, University of Glasgow, 1996. 30. Ge, M., M. Chin, and Q. Wang, An Extended Self-Organizing Map for Nonlinear System Identification, Proceedings of the IEEE Conference on Decision and Control, 1:1065-1070, 1999. 31. Ge, S.S. and C. Wang, Adaptive Control of Uncertain Chuas Circuit, IEEE Trans. Circuits and Systems I, 47(9):1397-1402, 2000. 32. Ge, S.S., C. Wang, and T.H. Lee, Adaptive Backstepping Control of a Class of Chaotic Systems, Int. Journal of Bifurcation and Chaos, 10(5):1149-1156, 2000. 33. Gomez, J.C. and E. Baeyens, Identification of Multivariable Hammerstein Systems Using Rational Orthonormal Bases, Proceedings of the IEEE Conference on Decision and Control, 1:2849-2854, 2000. 34. Gonzalez, J., R. Femat, J. Alvarez-Ramirez, R. Aguilar, and M. Barron, A Discrete Approach to the Control and Synchronization of a Class of Chaotic Oscillators, IEEE Trans. Circuits and Systems-I, 46(9):1139-1144, 1999. 35. He, X and H. Asada, A New Method for Identifying Orders of Input-Output Models for Nonlinear Dynamic Systems, Proceedings of the American Control Conference, pp.2520-2523, San Francisco, 1993. 36. Hui, S. and S.H. Zak, On Discrete-Time Variable Structure Sliding Mode Control, System & Control Letters, 38:283-288, 1999. 37. Ishigame, A., T. Furukawa, S. Kawamoto, and T. Taniguchi, Sliding Mode Controller Design Based on Fuzzy Inference for Nonlinear Systems, IEEE Trans. Industrial Electronics, 40(1):64-70, 1993. 38. Jacobs, R.A., M.I. Jordan, S.J. Nowlan, and G.E. Hinton, Adaptive Mixtures of Local Experts, Neural Computation, 3(1):79-87, 1991.

PAGE 124

112 39. Johansen, T.A. and B.A. Foss, Constructing NARMAX Models Using ARMAX Models, International Journal of Control, 58(5):1125-1153, 1993. 40. Kaiser, M., Time-Delay Neural Networks for Control, Proceedings of the IFAC Int. Symposium on Robot Control, pp.112-118, Capri, 1994. 41. Kantz, H. and T. Schreiber, Nonlinear Time Series Analysis, Cambridge University Press: Cambridge, UK, 1997. 42. Kennel, M.B., R. Brown, and H.D.I. Abarbanel, Determining Embedding Dimension for Phase-Space Reconstruction Using a Geometrical Construction, Physical Review A, 45:3403-3411, 1992. 43. Kim, B.S. and A.J. Calise, Nonlinear Flight Control Using Neural Networks, Journal of Guidance, Control, and Dynamics, 20(1):26-33, 1997. 44. Kohonen, T., Self-Organizing Maps, Springer-Verlag: New York, NY, 1995. 45. Lainiotis, D.G., Partitioning: A Unifying Framework for Adaptive Systems, Proceedings of the IEEE, 64:1182-1198, 1976. 46. Lawrence, S., A.C. Tsoi, and A.D. Back, Function Approximation with Neural Networks and Local Methods: Bias, Variance and Smoothness, Proceedings of the Australian Conference on Neural Networks, pp.16-21, Canberra, 1996. 47. Lee, C.H. and M.J. Chung, Gain-Scheduled State Feedback Control Design Technique for Flight Vehicles, IEEE Trans. Aerospace and Electronic Systems, 37(1):173-182, 2001. 48. Lee, P.M. and J.H. Oh, Improvements on VSS-Type Self-Tuning Control for a Tracking Controller, IEEE Trans. Industrial Electronics, 45(2):319-325, 1998. 49. Leontaritis, I.J. and S.A. Billings, Input-Output Parametric Models for Nonlinear Systems part I: Deterministic Nonlinear Systems, International Journal of Control, 41(2):303-328, 1985. 50. Lim, S and J.P. How, Modeling and Control for Switched Linear Parameter-Varying Missile Autopilot, IEEE Trans. Control Systems Technology, 11(6):830-838, 2003. H 51. Magill, D., Optimal Adaptive Estimation of Sample Stochastic Processes, IEEE Trans. Automatic Control, 10(4):434-439, 1965. 52. Martensson, B., Switching Function Adaptive Stabilization with Stability Margin , Proceedings of the IEEE Conference of Decision and Control, 2:1561-1562, 1989.

PAGE 125

113 53. Martinetz, T.M., S.G. Berkovich, and K.J. Schulten, Neural-Gas Network for Vector Quantization and its Application to Time-Series Prediction, IEEE Trans. Neural Networks, 4(4):558-569, 1993. 54. Maybeck, P.S. and D.L. Pogoda, Multiple Model Adaptive Controller for the STOL F-15 with Sensor/Actuator Failures, Proceedings of the IEEE Conference on Decision and Control, 2:1566-1572, 1989. 55. Maybeck, P.S. and R.D. Stevens, Reconfigurable Flight Control via Multiple Model Adaptive Control Methods, Proceedings of the IEEE Conference on Decision and Control, 6:3351-3356, 1990. 56. Middleton, R.H., G.C. Goodwin, D.J. Hill, and D.Q. Mayne, Design Issues in Adaptive Control, IEEE Trans. Automatic Control, 33(1):50-58, 1988. 57. Milosavljevic, C., General Conditions for the Existence of a Quasi-Sliding Mode on the Switching Hyperplane in Discrete Variable Structure Systems, Automation and Remote Control, 46(3):307-314, 1985. 58. Morse, A.S., Toward a Unified Theory of Parameter Adaptive Control-Tunability, IEEE Trans. Automatic Control, 35(9):1002-1012, 1990. 59. Morse, A.S., D.Q. Mayne, and G.C. Goodwin, Applications of Hysteresis Switching in Parameter Adaptive Control, IEEE Trans. Automatic Control, 37(9):1343-1354, 1992. 60. Moshou, D. and H. Ramon, Extended Self-Organizing Maps with Local Linear Mappings for Function Approximation and System Identification, Workshop on Self-Organizing Maps, pp.181-186, Finland, 1997. 61. Murray-Smith, R. and K.J. Hunt, Local Model Architectures for Nonlinear Modeling and Control, in Neural Network Engineering in Dynamic Control Systems (K.J. Hunt, G.R. Irwin, and K. Warwick, eds.), pp.61-82, Springer-Verlag: New York, NY, 1995. 62. Murray-Smith, R. and T.A. Johansen, Multiple Model Approaches to Modeling and Control, Taylor & Francis: London, UK, 1997. 63. Narendra, K.S. and C. Xiang, Adaptive Control of Discrete-Time Systems Using Multiple Models, IEEE Trans. Automatic Control, 45(9):1669-1686, 2000. 64. Narendra, K.S. and J. Balakrishnan, Adaptive Control Using Multiple Models, IEEE Trans. Automatic Control, 42(2):171-187, 1997. 65. Narendra, K.S. and K. Parthasarathy, Identification and Control of Dynamical Systems using Neural Networks, IEEE Trans. Neural Networks, 1(1):4-27, 1990.

PAGE 126

114 66. Narendra, K.S., J. Balakrishnan, and M.K. Ciliz, Adaptation and Learning Using Multiple Models, Switching, and Tuning, Control Systems Magazine, 15(3):37-51, 1995. 67. Nelles, O., Nonlinear System Identification, Springer-Verlag: New York, NY, 2001. 68. Packard, A., Gain Scheduling via Linear Fractional Transformations, Systems and Control Letters, 22(1):79-92, 1994. 69. Packard, N.H., J.P. Crutchfield, J. D. Farmer, and R. S. Shaw, Geometry from a Time Series, Physical Review Letters, 45:712-716, 1980. 70. Palizban, H.A., A.A. Safavi, and J.A. Romagnoli, A Nonlinear Control Design Approach Based on Multi-Linear Models, Proceedings of the American Control Conference, 5:3490-3494, 1997. 71. Patra, A. and H. Unbehauen, Nonlinear Modeling and Identification, Proceedings of the Int. Conference on Systems, Man and Cybernetics, 3:441-446, 1993. 72. Pieper, J.K. and K.R. Goheen, Discrete-Time Sliding Mode Control via Input-Output Models, Proceedings of the American Control Conference, pp.964-965, San Francisco, 1993. 73. Poznyak, A.S., W. Yu, and E.N. Sanchez, Identification and Control of Unknown Chaotic Systems via Dynamic Neural Networks, IEEE Trans. Circuits and Systems I, 46(12):1491-1495, 1999. 74. Principe, J.C. and L. Wang, Nonlinear Time Series Modeling with Self-Organizing Feature Maps, Proceedings of Neural Networks for Signal Processing, pp.11-20, Cambridge, MA, 1995. 75. Principe, J.C., L. Wang and M.A. Motter, Local Dynamic Modeling with Self-Organizing Maps and Applications to Nonlinear System Identification and Control, Proceedings of IEEE, 86(11):2240-2258, 1998. 76. Rhodes, C. and M. Morari, Determining the Model Order of Nonlinear Input/Output Systems Directly from the Data, Proceedings of the American Control Conference, 3:2190-2195, 1995. 77. Ritter, R., T. Martinetz, and K. Schulten, Topology-Conserving Maps for Learning Visuomotor Coordination, Neural Networks, 2:159-168, 1989. 78. Rivals, I. and L. Personnaz, Nonlinear Internal Model Control Using Neural Networks: Applications to Processes with Delay and Design Issues, IEEE Trans. Neural Networks, 11(1):80-90, 2000.

PAGE 127

115 79. Rugh, W.J., Analytical Framework for Gain Scheduling, IEEE Control Systems Magazine, 11(1):79-84, 1991. 80. Sales, K.R. and S.A. Billings, Self-Tuning Control of Nonlinear ARMAX Models, International Journal of Control, 51(4):753-769, 1990. 81. Sarpturk, S.Z., Y. Istefanopulos, and O. Kaynak, On the Stability of Discrete-Time Sliding Mode Control Systems, IEEE Trans. Automatic Control, 32(10):930-932, 1987. 82. Sbarbaro, D. and R. Murray-Smith, Self-Tuning Control of Non-linear Systems Using Gaussian Process Prior Models, DCS Technical report: TR-2003-143, University of Glasgow, 2003. 83. Shamma, J.S. and M. Athans, Gain Scheduling: Potential Hazards and Possible Remedies, IEEE Control Systems Magazine, 12(3):101-107, 1992. 84. Sidorowich, J.J., Modeling of Chaotic Time Series for Prediction, Interpolation, and Smoothing, Proceedings of the IEEE Int. Conference on Acoustics, Speech, and Signal Processing, 4:121-124, 1992. 85. Singer, A.C., G.W. Wornell, and A.V. Oppenheim, Codebook Prediction: A Nonlinear Signal Modeling Paradigm, Proceedings of the IEEE Int. Conference on Acoustics, Speech, and Signal Processing, 5:325-328, 1992. 86. Sira-Ramiraz, S., Non-Linear Discrete Variable Structure Systems in Quasi-Sliding Mode, International Journal of Control, 54:1171-1187, 1991. 87. Stark, J., D.S. Broomhead, M.E. Davies, and J. Huke, Takens Embedding Theorems for Forced and Stochastic Systems, Nonlinear Analysis: Theory, Methods & Applications, 30(8):5303-5314, 1997. 88. Takens, F., Detecting Strange Attractors in Turbulence, Dynamical Systems and Turbulence (Springer Lecture Notes in Mathematics), 898, pp.365-381, 1980. 89. Thampi, G., J.C. Principe, J. Cho, and M.A. Motter, Adaptive Inverse Control Using SOM based Multiple Models, Portuguese Conference on Automatic Control, pp.278-282, Aveiro, 2002. 90. Thampi, G., J.C. Principe, M.A. Motter, J. Cho, and J. Lan, Multiple Model Based Flight Control Design, Proceedings of the IEEE Int. Midwest Symposium on Circuits and Systems, 3:133-136, 2002. 91. Tim Sauer, Time Series Prediction by Using Delay Coordinate Embedding, Santa Fe Institute Studies in the Sciences of Complexity, Addison-Wesley: Reading, MA, 1993.

PAGE 128

116 92. Uduehi, D., A. Ordys, and M. Grimble, Predictive PID Controller Design, Proceedings of the IEEE Int. Conf. on Control Applications, pp.612-617, Mexico City, 2001. 93. Utkin, V., Variable Structure Systems with Sliding Modes, IEEE Trans. Automatic Control, 22(2):212-222, 1977. 94. Vesanto, J., Using the SOM and Local Models in Time-Series Prediction, Workshop on Self-Organizing Maps, pp.209-214, Finland, 1997. 95. Vu, K.M., Optimal Setting for Discrete PID Controllers, Proceedings of the IEE, 139(1):31-40, 1992. 96. Walker, D.M., N.B. Tufillaro, and P. Gross, Radial-Basis Models for Feedback Systems with Fading Memory, IEEE Trans. Circuits and Systems, 48(9):1147-1151, 2001. 97. Walter, J., H. Ritter and K. Schulten, Nonlinear Prediction with Self-organizing Maps, Proceedings of the Int. Joint Conference on Neural Networks, 3:589-594, 1990. 98. Wan, E.A. and A.A. Bogdanov, Model Predictive Neural Control with Applications to a 6 DoF Helicopter Model, Proceedings of the American Control Conference, 1:488-493, 2001. 99. Wang, C. and S.S. Ge, Adaptive Backstepping Control of Uncertain Lorenz System, Int. Journal of Bifurcation and Chaos, 11(4):1115-1119, 2001. 100. Wang, C. and S.S. Ge, Synchronization of Uncertain Chaotic Systems via Adaptive Backstepping, Int. Journal of Bifurcation and Chaos, 11(6):1743-1751, 2001. 101. Wang, J. and W.J. Rugh, Feedback Linearization Families for Nonlinear Systems, IEEE Trans. Automatic Control, 32(10):935-940, 1987. 102. Wood, G.D., P.J. Goddard, and K. Glover, Approximation of Linear Parameter-Varying Systems, Proceedings of the IEEE Conference on Decision and Control, 1:11-13, 1996. 103. Wu, C.W., T. Yang, and L.O. Chua, On Adaptive Synchronization and Control of Nonlinear Dynamical Systems, Int. Journal of Bifurcation and Chaos, 6(3):455-472, 1996. 104. Wu, F., A. Packard, and G. Balas, LPV Control Design for Pitch-axis Missile Autopilots, Proceedings of the IEEE Conference on Decision and Control, 1:188-191, 1995.

PAGE 129

117 105. Xianfeng, N., M. Verhaegen, A.J. Krijgsman, and H.B. Verbruggen, A New Method for Identification and Control of Nonlinear Dynamic Systems, Engineering Applications of Artificial Intelligence, 9(3):231-243, 1996. 106. Yang, T. and L.O. Chua, Generalized Synchronization of Chaos via Linear Transformations, Int. Journal of Bifurcation and Chaos, 9(1):215-219, 1999. 107. Yang, T., C. Yang, and L. Yang, Detailed Study of Adaptive Control of Chaotic Systems with Unknown Parameters, Dynamics and Control, 8(3):255-267, 1998. 108. Yu, X., Tracking Inherent Periodic Orbits in Chaotic Dynamic Systems via Adaptive Variable Time-Delayed Self Control, IEEE Trans. Circuits and Systems I, 46(11):1408-1411, 1999. 109. Zeng, Y and S.N. Singh, Adaptive Control of Chaos in Lorenz System, Dynamics and Control, 7(2):143-154, 1997. 110. Zheng, Q. and E. Zafiriou, Control-Relevant Identification of Volterra Series Models, Proceedings of the American Control Conference, 2:2050-2054, 1994.

PAGE 130

BIOGRAPHICAL SKETCH Jeongho Cho was born in Seoul, Korea, on January 5, 1973. He received his B.S. degree from Soonchunhyang University, Korea, majoring in control and instrumentation engineering in 1995. And he earned M.S. degrees in electrical engineering, with emphasis on systems and control, from Dongguk University, Korea, and the University of Florida in 1997 and 2001, respectively. Since 2001, he has been with the Computational NeuroEngineering Laboratory at the University of Florida to pursue his Ph.D. degree. His research interests include time-series prediction and nonlinear system identification, with applications to navigation and control. 118


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20110114_AAAAFQ INGEST_TIME 2011-01-15T02:25:54Z PACKAGE UFE0008180_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 39331 DFID F20110114_AADKMK ORIGIN DEPOSITOR PATH cho_j_Page_067.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
47e6c297ec7bb3fc7af328e5381e12a3
SHA-1
b1308140d9817e0950f3e9af6a9178136bcb9a91
21715 F20110114_AADKLV cho_j_Page_026thm.jpg
c663a6e652964f1e3818c36371944ea9
bbf256aa9b1e8e45e0aef2f61a86f883ff868ee8
74241 F20110114_AADKML cho_j_Page_092.QC.jpg
434d71d0cb1b5ae10c3fddaec386b9f8
bc6c6105ecac580abb8b30124bf3637b7b17e4f3
1729 F20110114_AADKLW cho_j_Page_026.txt
c4b6e49d92197af78f5a5ae3431bf48a
40e895d350fcd9381bb7e7823a0824b9aac8a09f
43865 F20110114_AADKNA cho_j_Page_063.pro
2a7963ffdf365e352e6693c38b293a41
25b35e49299c2e9b38cfcade84afa1ffdabfc841
225874 F20110114_AADKMM cho_j_Page_085.jpg
ed84845aa05d817deac2755f7519555b
096a434e7553fad4fc886f1c3f628e3b3567fe6b
2954 F20110114_AADKLX cho_j_Page_009.txt
896f5f2dfc6441e2bc9b246f21be67f9
972b85a16c55ea69b263f3e7b8c2f25c28bb3274
1232 F20110114_AADKNB cho_j_Page_101.txt
3a3a4bbc4c571c81236cf1f0eb09f556
61c1eedcd1d1d6cc133bc57c4355b5fd55b58e3e
1053954 F20110114_AADKMN cho_j_Page_122.tif
d48ebeb12e73d3ed8f9e74286200dab7
97119b44fdbf97d55284e9b3dab7a073d6b24ee0
139406 F20110114_AADKLY cho_j_Page_039.jpg
473191d5497d0bd2c4fcf8d633e666c0
014d7418e986426c53056b26ec45c6578bd49333
1814 F20110114_AADKNC cho_j_Page_054.txt
864d792cd6995b5fd4c72de287d7395e
0eda9b4b610adc581ac8e6c429729126e42e74ca
25271604 F20110114_AADKLZ cho_j_Page_086.tif
69c956c12c418fddbb9596528035ad61
9b028dd68baa2569dad472b48ad595e63b8dfc7c
43318 F20110114_AADKND cho_j_Page_085.pro
ae5bf77d56c9c3fc225ea1e18421948a
9545893e6d8c40a45d1f550eedcfb37f98a9e0a0
48013 F20110114_AADKMO cho_j_Page_096thm.jpg
3bf58244ead7ce17423b9c3f0b7b7e15
a7d78d4947dc8e1808fbaa1043e9d7ff7e9c8362
F20110114_AADKNE cho_j_Page_028.tif
89b6e54d4c1268bfbf16b04d07d5771f
5c04ae0b054053b3eb9649125330ca51fecc3f4c
1382 F20110114_AADKMP cho_j_Page_053.txt
429711320f20129c8d883e6f03c15e18
1930a59927c6cf341664f9433358f4c06c3ba80c
92040 F20110114_AADKNF cho_j_Page_063.jp2
b5083c692aba8801c7136767dfe0fbd2
b584d483beae81ad6c6fa49bfa004cfc64279112
69846 F20110114_AADKMQ cho_j_Page_120.jpg
1b194d5941b660dd364f8a2f0a2de877
d83d70aaf671fb781ae909be610e0f96b560e35d
41381 F20110114_AADKNG cho_j_Page_118.QC.jpg
64d2011d2f46fa4cb4b05bb27515ea10
cdd5ef33bb575fb7a17b5dbfa1ea276a4e883095
42061 F20110114_AADKMR cho_j_Page_074.pro
a944c6534269054850ace7e2f934917e
ca2a8d472e1e71a1a0af1c10faa5bda8c2e3d384
32316 F20110114_AADKNH cho_j_Page_073.pro
af14634bfa655f762ab8c4fe1c5858b7
0e303ffb49cdb4580a71ae5b9bd8406c770ae9b9
F20110114_AADKMS cho_j_Page_032.tif
926de933d5090c57a7f8de201f6570a2
e00e789b1c9638bf04bb8ca3b53789c2dc81a397
40388 F20110114_AADKNI cho_j_Page_103.pro
d4f6f9ff76d7be3dca734ab85ffb7605
57a25cf938f6957a4a9ded423e6d9d9c8abc8f04
1396 F20110114_AADKMT cho_j_Page_089.txt
2d17a8ee8366f7ff7e948bba73e0d01b
8b6c2f2361711d484143b9e18621e6afd6660613
1843 F20110114_AADKNJ cho_j_Page_020.txt
88ffc20382d6c41eb2586c099e48d394
006f31394f39b2f4a4e1e535615211ef2ce570e5
88375 F20110114_AADKMU cho_j_Page_103.QC.jpg
051515922c8ce63bf8f6f986b7ed5e49
8ffcd636206432e2017eb60846114345e416e2c9
62151 F20110114_AADKNK cho_j_Page_062.QC.jpg
bc6a185e60042931f12d0c244686df06
1c101ad3de044eafe8f90e2a3a628b84c9104995
2047 F20110114_AADKMV cho_j_Page_049.txt
85f8443997957496e2b5cd5392fd13ba
47e4f0370a13ed480f5b1810d4737247cca804df
91302 F20110114_AADKNL cho_j_Page_074.jp2
1411ab61fbf0fdb9691945d8cf220854
e1a0bdb0633dabda2dccdcd0593fe4a4120429fc
1431 F20110114_AADKMW cho_j_Page_091.txt
67f08763d10357ab66a2af9990a41168
2870f0c40e341eb8e6494c9299c66a3a68233e52
82670 F20110114_AADKNM cho_j_Page_124.QC.jpg
2c0dfb7cc8f5d8e1d9ad26ea9ea70c8b
10aab83e09d1d3ec27adb3b57c0fcd9dc4911bfb
3215 F20110114_AADKMX cho_j_Page_005.txt
99da8ab8c202ab8f83497a6fa6ebd65f
68b8d40e95caaab50080e926a4675652e644433e
187822 F20110114_AADKOA cho_j_Page_113.jpg
034370fbe7edf7b8975721784c559588
36e79f4603ae4000ea36bb3bbe40c6dfe6364fcb
2081 F20110114_AADKNN cho_j_Page_105.txt
05cf41ca1a61fe305c207849ec4dc18d
c6113776e9c37d5222e4da6651d676d5f2c8258a
18601 F20110114_AADKMY cho_j_Page_120.pro
b1ea4046437f5ba4e9c61e5157e28750
e106bb187ae56bd431561d13c77188c377dba301
193739 F20110114_AADKOB cho_j_Page_050.jpg
8ce36773eb45963793baebc283a19442
15465b59b47a83be9e36c5cb10849a16e4aa4e52
17091 F20110114_AADKNO cho_j_Page_065thm.jpg
54c53232939e3b2c9760fb39ae9a9548
6c32f191186279a160f9afc9b1a22d6facf93b6b
F20110114_AADKMZ cho_j_Page_059.tif
0b9c1586ef2fe7882f16f79aff0497e9
6d533166e81e9a6c87d6b6039dc3660ae9c5f1dd
51714 F20110114_AADKOC cho_j_Page_016.pro
c92bcef440dc59adf9258cc3d1980b73
00c2ba346589bec19e98665c507a7146dc3e847f
2442 F20110114_AADKOD cho_j_Page_010.txt
e30705b6fe9fddeb99c0625effc40470
982204871d3cdeb7790bd9a4f8f813ede22e8c7f
79041 F20110114_AADKNP cho_j_Page_055.QC.jpg
a97a0c3b6cab2f1be597e70790ea4ac7
3424628300b74f9dbbe25dc7eb713c6908a9c1ec
184545 F20110114_AADKOE cho_j_Page_076.jpg
4d5a06ca67c654095e683205700858b1
92771587b39d67bb7e5dec5e7d961b4748c7c5a5
35712 F20110114_AADKNQ cho_j_Page_040.pro
192e7f6954200d3946b0deb71e0e80f6
2fd0779aa86be0a992dc83b988bf416c427a3614
93467 F20110114_AADKOF cho_j_Page_102.QC.jpg
6f79ca86fec1af22ba918c1582f0ec87
cebe406a941fc33bcba91a1c53dc81362b6ff433
F20110114_AADKNR cho_j_Page_071.tif
892835b05d858ab8b2f9b3bb402a9abf
4ef2019d0f051499ac521db5fb5b4d2dfb24daf8
187796 F20110114_AADKOG cho_j_Page_060.jpg
0602885abe4d1b80173b5192b375cdd0
f0cc9dc2afe490287e7d3ed8f37bd7524468d9bc
27079 F20110114_AADKNS cho_j_Page_052thm.jpg
46513798222e54d6a906f583d2f082ec
5285943d2fdbf7b42dd882ea76e68d8ecfb4858e
1526 F20110114_AADKOH cho_j_Page_068.txt
13ea3802b90feb9c1ccf816deeff6074
30e4a337c5c8777069d590a4fe054319a4125868
1051984 F20110114_AADKNT cho_j_Page_010.jp2
d17541f97dabe6eaff7b78565919a49f
9b8e7cba5a48ce3ab533fd834779ffedd533df18
993099 F20110114_AADKOI cho_j_Page_083.jp2
5bee63323ffd526abdcc90a8f7f73c1d
4b8bda27aafa9a501bc5df60e14b20b392cdd4ba
27277 F20110114_AADKNU cho_j_Page_058.pro
613ecf3cbe4e827c2a0bdc9026362b4f
9b7ab2edb3b6db67f1421c36d5bb0696721f05e4
98215 F20110114_AADKOJ cho_j_Page_020.jp2
cb31a536670446f1a3bdb232217122ec
2ad61df0f7d6ab613bd8bcbf208e2f60ff7593c1
130968 F20110114_AADKNV cho_j_Page_125.jp2
e1c4d41415f41ddb7488f1d9cee99a9c
1c7b3f0ccf81eb17d086c5dc2182ff42430ee38c
44994 F20110114_AADKOK cho_j_Page_087.pro
efae8b152005759c83d0bc33f3058694
08537c87291158537786cc9e5f4e007270aff0ef
8423998 F20110114_AADKNW cho_j_Page_058.tif
e3fd2a54975236a8f904eb4183c64285
40e52b1ee41cf44af6a4a60577968ef912abbd29
56974 F20110114_AADKOL cho_j_Page_029.QC.jpg
ffc3f6edc9a99348a583d9e6ab998df6
3ce7116402ae845246a7aaa72ea54a491dc78d4d
1051967 F20110114_AADKNX cho_j_Page_006.jp2
61750114ce2d61289df07cbb530fd109
5fb6d5a6e7b688ac28459492b3b69349454012a5
F20110114_AADKPA cho_j_Page_106.tif
fee49389a20450c7b80274399b1393cf
1d233e074e59fdde02eab0c8c5960047611f55db
F20110114_AADKOM cho_j_Page_041.tif
a8b08c834329450038a04d15ad63ae72
c072f03517245c488304809c1e414d16ae6adace
F20110114_AADKNY cho_j_Page_078.tif
95e7b69f8c34c836e30daf45e28168b2
f0ecb6b3927c305405f5ae0d5fc62a933721b19d
F20110114_AADKPB cho_j_Page_107.tif
2659bcec631bbd17a66f685fdd1eff13
a5ce0bbcf0cb16ea2f6b914594773886ac341eb8
128750 F20110114_AADKON cho_j_Page_129.jpg
8008ca21512008a0591795ca840129a2
9209a43913a94cd45342832efc9ef04dacd13b97
349613 F20110114_AADKNZ cho_j_Page_009.jpg
8a0aa25f19ade3a3949de5b52fe5808d
d899089874553d5fa7f6dbe8f7c9adf1f5cd7521
170178 F20110114_AADKPC cho_j_Page_095.jpg
55edcbb4fcd3ad0e01ec147042e8fe26
35dbee0c46d5645f57eddf67afe13a03705d40e1
24482 F20110114_AADKOO cho_j_Page_042thm.jpg
9b4fb70cba7e95d0ec57ad4dda827c32
599acf3a1485cd008d18f559cdf93d7d832044b0
60778 F20110114_AADKPD cho_j_Page_109.QC.jpg
4eeae431a3deab08dba8f5fbaf475567
2e0effcba99931c2364851af62dfaefbb48ff5c4
F20110114_AADKOP cho_j_Page_064.tif
1ba8f269b167d6522017fdaf36f484d8
f4d1437c7106bba113beea75e06418bd36c65aa4
F20110114_AADKPE cho_j_Page_007.tif
01234a677e6edfe21e0c94c416ddba4d
e0f77d69333a2f1c197a2a0e681ededc3812b49a
44942 F20110114_AADKPF cho_j_Page_072thm.jpg
1dd22cbe01b7c10bf4567b9347abd615
acdbbc512308f76fad00c798323b88e1bd2790ed
69688 F20110114_AADKOQ cho_j_Page_043.QC.jpg
ef48f7991cfb42f8e53c66bfd2cebed9
38a78183b2099cb0b2ff738d61c743008f4eb05b
36414 F20110114_AADKPG cho_j_Page_078.pro
475ded23b8e5beb683811e5a9916d913
ae1d27966e4bdb284bf11a8ca7bb548b7a7769e7
1362 F20110114_AADKOR cho_j_Page_027.txt
e3529cdc192114dedc6fafc04d888f9b
30d7da6978362b1e147b67a7d4bbbbf94a3413b4
49202 F20110114_AADKPH cho_j_Page_037.pro
133e0020d740544052be19a7b10943df
3cea15d3da865803e8a5022dcaeef243a1c9515e
88027 F20110114_AADKOS cho_j_Page_026.jp2
13583e96bc9277a36ab53bf87bdb56eb
9999852540b5631b8860517bb29b65a8c5161ee9
223497 F20110114_AADKPI cho_j_Page_115.jpg
fbd90ba67afe0e06ba26cbc4b6e3f8c1
b8f5fe72bda83a34a96d5e2bdb25f6daaa80f42c
89808 F20110114_AADKOT cho_j_Page_119.jpg
165db278f03f201a68a1e8eb575fe523
8a2f05057734ec5649ff5d3a875de1c214bcc244
1450 F20110114_AADKPJ cho_j_Page_051.txt
5143e64accaec4226ce41e1b535b53ce
73bdbb2f9f41a661678b69b98852532ccdf46e39
941546 F20110114_AADKOU cho_j_Page_048.jp2
283feeda026e140c49563d0012b6835e
aa424433cc3478fad8d50129a0cb6ec9092055f1
F20110114_AADKPK cho_j_Page_085.tif
2aa3ac4b44f4acfef4b4a446cfc5d27f
5cf28808974e60c465e8597011eafb86b9bc4c17
47539 F20110114_AADKOV cho_j_Page_102thm.jpg
34f4351ee5487c5e7b8281b66d1db861
e90b39490d9a59d15878c56b03464b38a375c640
49113 F20110114_AADKPL cho_j_Page_078thm.jpg
b7e30fcdf2b701f8b43442ed1ed01aad
7ad8787fd9bbcb710c4b94078e5935932b7e859e
1740 F20110114_AADKOW cho_j_Page_072.txt
e49ef289040e7fdbf1a8580eb815762c
99ad45dbfbb1a21dc7cf7ac7291c9fa0621c2e52
21188 F20110114_AADKQA cho_j_Page_067thm.jpg
2056b1d73cb93296ee3419c6ced00074
09fbf8e96a49645ee55f5fef0276ad70a087145b
46655 F20110114_AADKPM cho_j_Page_094thm.jpg
19a5fce308e99042e5099b04a72f81da
0a56c0269ceb315eba686c4a3b2cbfc92989ff2d
882521 F20110114_AADKOX cho_j_Page_088.jp2
39676f893b2e117f8e337d1ce71b9668
63d8265ea7ca972a8f5af246641229729c822912
45091 F20110114_AADKQB cho_j_Page_090thm.jpg
647960d280935d5aff19d70a1d5bbda0
9d45e9f6f368056b61fd7c0ed0e77d376c966230
1374 F20110114_AADKPN cho_j_Page_088.txt
4e567b658890c2cd2784b3d0dc718b6a
3d67f3dfc61d45886188e90abf8229052cdfae0a
1337 F20110114_AADKOY cho_j_Page_104.txt
dfbf4bd4747ccf6f11dde0822451c8fe
2a6ab8a3c87ba7fbbef5fda3e873fdbaf48a9df9
83695 F20110114_AADKPO cho_j_Page_104.QC.jpg
9ff05880e40d992726ba4c855146d66f
64455808d0f6435512ac69d23db3db3360b03749
927101 F20110114_AADKOZ cho_j_Page_112.jp2
86cb4d5ecbe4ab4eef2552fd4636908d
0137b16a1b1b23431cab6aa8746abad8c18e0ff0
85004 F20110114_AADKQC cho_j_Page_072.QC.jpg
87548dd1c6eafc1c0b97e758a461ccb2
350f04c8c072ade07bdad662c8c375a912d848a4
61658 F20110114_AADKPP cho_j_Page_028.QC.jpg
fb7ddd24c00278ca97f62b70121874f2
7ba749fc9d9737202544508d8226291594a7c218
F20110114_AADKQD cho_j_Page_088.tif
6088aebc94f5440902e6153de7ccab95
effe7558cd36b26bca9088012b78193214702042
40460 F20110114_AADKPQ cho_j_Page_026.pro
7185bf79dfc0e35c5e5da2e323f6e924
772de1426ba00b4b2aad94363e9ae52229a0452e
1824 F20110114_AADKQE cho_j_Page_070.txt
f2e49944009369a71e6aea64c6dc18c6
1a4f0c55d339fac30d2cc7ffb23243dad25ef74d
1665 F20110114_AADKQF cho_j_Page_069.txt
f2f8c49ecbf7b995613144b2ba6b2d38
56d6ca69efd1937d68e7ef43cbd7d91ec05c672d
83683 F20110114_AADKPR cho_j_Page_018.QC.jpg
6282ce1aa719ed6f58aa1a0d88cba654
28af7efc9a8e675eda3ccc8d05bf4c79743c5a6b
45586 F20110114_AADKQG cho_j_Page_103thm.jpg
a2705cbf65a2683b21fef91188901de8
9b976d6c04f47b1ffb270140bb67e79497fdb661
21742 F20110114_AADKPS cho_j_Page_029thm.jpg
fcb4ea231ded2a10b11e6a35d545e6f0
82b0c2301eef35c0a22fee7d347337f15a0eeb8f
F20110114_AADKQH cho_j_Page_065.tif
7531e9020940892d500b2d5503b24fc7
ef81b2e6cea3e82c8e087e4c5f9bd432aebf9752
F20110114_AADKPT cho_j_Page_014.tif
b9adf2a0a5674f63642a4b0233f6bf10
ee6642618d71996046e87f329a1b2371c8dc9129
2111 F20110114_AADKQI cho_j_Page_014.txt
f515bcda58a641e216e4ddcb9c913098
eae5ee2a19852d2616f29aeeb0c20e4c5e657f64
47576 F20110114_AADKPU cho_j_Page_108.pro
a7919eb4924c88a0f0721b5f3e01c9ed
97fdd0b93ec60b7f8dbbf220c5db7dbde544f6d5
75056 F20110114_AADKQJ cho_j_Page_031.QC.jpg
703b53f295c76079c33d77d30befd4c0
89a8a1d66a7b2e0ee6ae637cb4b859c81b853de6
159667 F20110114_AADKPV cho_j_Page_011.jpg
079841f11dc7369dd4e57f83c4009826
5c8f127d270536fa4d9b4faa8000a7cd8294473b
48528 F20110114_AADKQK cho_j_Page_085thm.jpg
60eb4e974a62cb5474a770ee13a5f30f
04498cbd8959411674a35d87a1de56905a4012d0
F20110114_AADKPW cho_j_Page_114.tif
74c9e67d5faeec3c5d3bdc44786cd6f4
6bf083cd4d21bbf31c44ce7d6235e738eeff894b
45104 F20110114_AADKQL cho_j_Page_075thm.jpg
70beb708754f53672b8394d15a23d2ab
6fa6ffc6c7bbb69fea26488e76d6d9bd8bfc53aa
115308 F20110114_AADKPX cho_j_Page_128.jp2
49d5fdc103dcf4496f43ea5c5fba4449
2fcc631fae7e1377cb427e7d10d6f9bb03fb6b4a
53498 F20110114_AADKRA cho_j_Page_010thm.jpg
fc0e6b9d959b7cc0fc0ab9ff1eb2bc7f
cfd135394083d443451ffc3eb27e1cc89ca53ae8
72691 F20110114_AADKQM cho_j_Page_070.QC.jpg
cdc2d75dc863a7b9d9ffb792f81f9467
014e372efb30126ab239b99804d22a48e1dd4fbc
45085 F20110114_AADKPY cho_j_Page_047.pro
367ff930ac0225a8b4d0d56cb8be8ca1
e6b645c3cfb1be147dcf475595e0391f3541fb62
F20110114_AADKRB cho_j_Page_045.tif
8f2e6f2064385b89ac53c4c1e903fa2a
858e46a52d509da9ddcbe308b3875426d3d0cdf6
45890 F20110114_AADKQN cho_j_Page_097thm.jpg
57b1ea76cd3436c55ba804c41c3fc549
398f6bb5dfda4eabcd4a777f0328f1f120c3f298
83411 F20110114_AADKPZ cho_j_Page_130.jpg
e802e86cf17576b9e87b3ab479a28872
a344960aac5a050d276736b92da55d679be9b9be
41781 F20110114_AADKRC cho_j_Page_130.jp2
13355269f08c895bf46c4401a2e7d4dc
59fe2bd93cd9dc3880d93841f2490bdfd2113aa9
86723 F20110114_AADKQO cho_j_Page_075.QC.jpg
9a402cdfa0437828f3a0cf3a0a27789b
59ecce7ce81bf1dbe95556317b18c8901e66741f
190807 F20110114_AADKRD cho_j_Page_099.jpg
5e0cffa6b79169e752124ced7d9c0844
3c44add668b4bb9b08e15bf5547cae927d04529b
96650 F20110114_AADKQP cho_j_Page_060.jp2
8a420774883d6215a703465f4dc3063c
a96f8cb055838491e769b8f405c4722fd347640d
128339 F20110114_AADKRE cho_j_Page_127.jp2
2019f0af8d1fe7133cdd4258ab271a8f
2dcfd6f76195bdfa46d022f2aa7bfa3a9ef89551
37182 F20110114_AADKQQ cho_j_Page_109.pro
620a321cf915145c7fc8b5103062a18f
5699c8a360af290ed18a166c0366c665a4d23ebe
505 F20110114_AADKRF cho_j_Page_111.txt
cb1e92092bda6d02ad3bfc3c3519686e
833be92bcba6c09f919c24a87b272542a82a8b6c
1103 F20110114_AADKQR cho_j_Page_129.txt
4b32b067cd446bc15348b654a59a5399
f389c7b524807c7965dd5db644807e83a64be393
157498 F20110114_AADKRG cho_j_Page_025.jpg
8df5a642bb54cb35809780356fbc1e28
e173d0e965300247832c553ee655a608f23f8de4
77689 F20110114_AADKRH cho_j_Page_050.QC.jpg
b1a92727bb5534042623488a5825a995
a5aad7644b08b512322fff2d5b8168a47ce8dc29
22162 F20110114_AADKQS cho_j_Page_119.pro
2a494d34ca713e5c04cf2dda413be143
5b2b985b93b0284b57db948c45babd1193e69998
72388 F20110114_AADKRI cho_j_Page_034.QC.jpg
ab4508dd34dc587051820f1a51b89b3c
a11682b96a01a1d0a85edac2a47b0ff85b4a4342
47656 F20110114_AADKQT cho_j_Page_008thm.jpg
3e50d24c5c9c9b73117b16783caaa7d1
81650367020ec182a51fb76f85bd2ee2c238d0cb
83530 F20110114_AADKRJ cho_j_Page_100.QC.jpg
467bfa7bdb1aee4c48eeb5861ed6b986
a21878b75147155577f90190fb19a5da030827f3
194506 F20110114_AADKQU cho_j_Page_020.jpg
91babbfa3e2258098497e481df939f4f
defceb43322a05675762aa83323abd58199fcafa
1791 F20110114_AADKRK cho_j_Page_013.txt
47bb7388115bd121963f6cad1dfd5202
f627212348d300e032a41ebe6751ad9aafee277d
F20110114_AADKQV cho_j_Page_027.tif
0d9da90f4d564a4ef14d786a37d56636
b70d44df99d1d33e3c38ac8c53e564a275724a7b
64521 F20110114_AADKRL cho_j_Page_064.QC.jpg
d35d16cce0885c4e6da4c4e46a8df24f
616d7090567380084bb0e6750f8990e5f4096c33
66110 F20110114_AADKQW cho_j_Page_033.QC.jpg
ae71b8e6a479c338f8da1cc63d621d07
af4556ae57990751373a0fd64c0229217c322886
44492 F20110114_AADKSA cho_j_Page_020.pro
cfc84605a51978d78f711b8c05b65838
c226f2cfc678ad863c0a8a73c085b05b0407aafe
122382 F20110114_AADKRM cho_j_Page_124.jp2
f3a1ce1b85a3d6d8417bffb24155408e
5ba9839c5e6e5bba2b872e87c155eaa99c63fb89
2113 F20110114_AADKQX cho_j_Page_077.txt
8f955808bb62cebae5f253af2eebfe9f
29363e4d76152974ee165c0c45fc25e854eb9482
38107 F20110114_AADKSB cho_j_Page_021.pro
f4aaa30462597c135268166aef5286ec
73bc166bfef794adfc962ad521b30cc81c15ef46
29693 F20110114_AADKRN cho_j_Page_046thm.jpg
7a55409a1906831bbad3bde3c16c3287
c34fedc5fc8036307b0a5c59a7c5fb5f2d7ad10c
20672 F20110114_AADKQY cho_j_Page_021thm.jpg
b5b0b64696b29ddd2436786723fb6410
08b789a831c512bf131e92e4ab51c18f6af9f756
102297 F20110114_AADKSC cho_j_Page_023.jp2
72b909934ea784b84c5e9baa2e8ccb9d
a5e15566c66e248848de51ddb3418d4618e8eb1d
65650 F20110114_AADKRO cho_j_Page_004.QC.jpg
2330b18f96f469defa2170eb9b47b2c9
3c87fc338f36041464dd4738c070c6fceeaf5cae
35439 F20110114_AADKQZ cho_j_Page_120.jp2
b298e287b9a64823b6b3ae8a4c793a07
89710831aaca1798664765e8e80eb5f56d54500a
105557 F20110114_AADKSD cho_j_Page_008.QC.jpg
0412c5a8d5b63cb79af4d2efbb60dfd7
1137cca86bcf88aa8935eecb676009c7516da2d3
90400 F20110114_AADKRP cho_j_Page_116.QC.jpg
cc45d7bcef9cbbd67dc36b3ce56fd227
29002d193b486063a2833a03952277413da8d5d8
1051980 F20110114_AADKSE cho_j_Page_008.jp2
3f7961a0235f4fd5249643b83fec6611
63dd858adb98f612b6fac2c96ba843aa997abdbc
76063 F20110114_AADKRQ cho_j_Page_082.QC.jpg
7c22d9fa7e4c8316318bea2f423ffe02
060a788d6fbf3749ede6a88f5f72bad33520a122
72302 F20110114_AADKSF cho_j_Page_061.QC.jpg
8a2ea4f8de3d9ecabaa833536490082a
23077b8611644310512220de782afb447203d1f0
286497 F20110114_AADKRR cho_j_Page_116.jpg
63fa589d1b5fb3cc20716767cdafde52
d75bfe5836b8a0b24fbaded78763db8388df30c1
23419 F20110114_AADKSG cho_j_Page_066thm.jpg
bdad8265e4b66bd6762adf8c62fa574d
723c298a3f352c292695e0d256b1a7843de8d42f
731817 F20110114_AADKRS cho_j_Page_022.jp2
f5a492efbb45b9f3c6aa25bc4baa091a
e6dc0b4e75daa59377455cb9617355305d35c95a
36834 F20110114_AADKSH cho_j_Page_102.pro
05122e5062c51edaf680197bd9e4ec61
7cbb59691bbe59229e89e3fcd3e18e1fa0acf956
42272 F20110114_AADKSI cho_j_Page_060.pro
cd9acc28171a12eb592a6355cba8ae86
aa464dd88519ad39c4c6597fa5b6c0fb1d60c37c
192480 F20110114_AADKRT cho_j_Page_035.jpg
d04807471cc6f8ecbc37774693e11b21
68c17b0d1f75ae16b47229a72a070bda2ac1b1cb
33176 F20110114_AADKSJ cho_j_Page_106.pro
4742a80c691a7e6064bddc078356a7fa
6cd912497cb669ff27e6790b5419c3290758183b
1632 F20110114_AADKRU cho_j_Page_078.txt
256eb07797b96b9dd5ab61881a2cc8a4
99c8f710a4122ba02d08af61d63641dbdd22a301
1800 F20110114_AADKSK cho_j_Page_074.txt
31f378d6b5642a9409a09b24a0e85f16
b41741f7ee8daff043ce43cb9eba9b6fb08e7ff6
25739 F20110114_AADKRV cho_j_Page_044thm.jpg
3b6459cde10a4c38e49b65a9373499c1
544137da12e0b166bf79c29cfb7e44ffb85fa4ea
F20110114_AADKSL cho_j_Page_124.tif
5f875fc47e5882864202d6db30af3627
f3b14435a8e5b9d243fda19fb5ff58ba0eb16576
1692 F20110114_AADKRW cho_j_Page_048.txt
0c410fe575751acefbc770125b9d3a5a
6d3307bb64fc2768af3fdedca515e6f483cba4c5
101465 F20110114_AADKSM cho_j_Page_049.jp2
00b2f074944fe7c306190b70c595fa82
13b7f529647897fede634ba75bd54cd3be56e2ce
198361 F20110114_AADKRX cho_j_Page_049.jpg
fc65f139315e6a313105d174a4c1c72c
cc8d09a0f6db23bfe1b63a5ddaa6e71332571efe
F20110114_AADKTA cho_j_Page_113.tif
f3579dec2a4395d6a5c93ddb64033c55
7f05c4335b1b470ffdc888a6fa27214143878a84
255527 F20110114_AADKSN cho_j_Page_008.jpg
6439ec5a75f64dd5299af83287b798c2
b240026ad15a3c0f518bba79211cb7d7566e6280
40132 F20110114_AADKRY cho_j_Page_062.pro
cb2e02d864027dddf2232e7570e73050
990854befa8ee98ea269be27f99f6993ad041c86
44627 F20110114_AADKTB cho_j_Page_032thm.jpg
b62971307d6240160d0d667c7264617f
31be1e5708a2a02796db3e075825c7225dc1ade0
35497 F20110114_AADKSO cho_j_Page_034.pro
dfa12e97dddfc239a3a0c36aaa5883d7
5ed5ae7c89946fccea883b19af42de357810d1cf
56032 F20110114_AADKRZ cho_j_Page_039.QC.jpg
39b382b86c1838c1a462a4a1075fc20c
cf05847e5c4d544531503f31c768d198654191e2
55580 F20110114_AADKTC cho_j_Page_009thm.jpg
42bd48aaf8f3ebdf866a7b43b44649cc
f493a981b4f946cac88a28273a83dc2a62a3ee64
F20110114_AADKSP cho_j_Page_117.tif
a53f234ba828a55514e51a2956ee4388
042a34108d28ad1085c93335ce5a095397aa5d77
62084 F20110114_AADKTD cho_j_Page_110.QC.jpg
7a996d4e94192d47c9cd3e3c740fb67b
ff7f7e6b92b3da6e8a7b418c15f813f729878cd5
84783 F20110114_AADKSQ cho_j_Page_115.QC.jpg
c94b27988750e1348b181d263e7a6871
94afc26e324e47620fc360184cd94f561d8fd165
66313 F20110114_AADKTE cho_j_Page_030.QC.jpg
257323f2f78784bd7054ae6b5e65bbda
ed3021fd1c92293081b22d820a3c45c98a231eea
93547 F20110114_AADKSR cho_j_Page_122.QC.jpg
d95955380a560691e0ded75db8e1fcb2
f8e016896f82800377a0bdc85477d34e56e129f5
F20110114_AADKTF cho_j_Page_057.tif
85a9a10ef533dda2a499b34677ccf041
366f7e790983362f7c4590081b3ff1c5d4029500
1555 F20110114_AADKSS cho_j_Page_098.txt
9d2bff73c9cdca2d6916057941f2be95
58307ca6b6bb51a8bd596234c327f58a1903acaf
67229 F20110114_AADKTG cho_j_Page_063.QC.jpg
04cc85f40cde2591ede37575f7d39ea3
38a6cb4abc0dce1380fdfb39f8b8c94c1c89fe45
F20110114_AADKST cho_j_Page_020.tif
8576f9e46744186922efb9e578f26edd
21a2c86d16e4a4875c79a755340529036b0c95d6
29923 F20110114_AADKTH cho_j_Page_088.pro
493011ee82e478f576b45dbd8deb2c2a
d5e5cad368fb3aeb6e16d51638dc9bd00f9d8f9e
42763 F20110114_AADKTI cho_j_Page_033.pro
be8d02462930188da877468eebdaa753
c8a0791b6b66d4e3722c17c097ffd956e9a9ca98
49063 F20110114_AADKSU cho_j_Page_080thm.jpg
99ceddc963a9df9a29fc93d7bfc5013d
d1469ea8b0e0928e766a11580ee41eb018dd9830
112417 F20110114_AADKTJ cho_j_Page_016.jp2
200750c45b1bb85e46991b2a06d1fc21
25ee8becff20ce234f2d6fce821e94835a57ddfe
215753 F20110114_AADKSV cho_j_Page_073.jpg
145f9d0ad48e44350446dabf30c3ff0f
c9c702c3435c9fe485d7b9207da0fd70d1546ac8
45768 F20110114_AADKTK cho_j_Page_007thm.jpg
3edc88ad34d549989331c213d8903e53
3d61908d3a5f24fa1a470475f71fcbda92834289
443 F20110114_AADKSW cho_j_Page_001.txt
abc3adb9c6084ffbb77ac8db5cd07bd3
c2ffc30b5cefdac284eab778e53659f6ac0af786
53812 F20110114_AADKTL cho_j_Page_040.QC.jpg
ee7cda08dd17c477dfe26a12e34a45c8
371f7f0ac749de00be44ad746ff140caf597b403
25588 F20110114_AADKSX cho_j_Page_049thm.jpg
7230f60e83b9a3df2ac1a879b24a37b8
73d3e3fb57c82136f357b16b7feb4ad30fb3ce53
74362 F20110114_AADKUA cho_j_Page_009.pro
461b28c634d7bc66614be8d4259a5bbb
bb929c1da0b4099ee06d4d2a55438d54418f6939
102653 F20110114_AADKTM cho_j_Page_045.jp2
3552dcc78d73427014c11656f6069d00
9da5842ca88004343c611c5f6e0a002d34c55933
37709 F20110114_AADKSY cho_j_Page_072.pro
814dcb23126850f6769a0e153b8847b3
5bf0f0dda68af83528e25448380c33f337cd329d
781907 F20110114_AADKUB cho_j_Page_072.jp2
f36bd062d4e068a007e33fe8986fe47b
89190d8beade8758203d9e30903ec0d2e0684c7a
1881 F20110114_AADKTN cho_j_Page_067.txt
e4de60c8261297113de9028b4c77ac60
d81f77aa7614dd89e992dbc791677b5b415c771a
1022906 F20110114_AADKSZ cho_j_Page_094.jp2
7e36d17b68936119a936d3cf9e4a0c6e
f3730760991123f940c90df903dcd6989439839d
F20110114_AADKUC cho_j_Page_072.tif
1b4f0452a84cc58b7a055aa8766204c8
7c5dc891b02b2bb83522bc0b81e53f6d5da90f06
30481 F20110114_AADKTO cho_j_Page_056.pro
9173470f129c201037f5f7f9963100ba
c79f3b45fded51863545f21e93e30519981e3dc0
194053 F20110114_AADKUD cho_j_Page_097.jpg
8e1f5131c0b05172d0de7d482bcfa80b
b8af3a70a1d061727267a9c0861c2e18aa282bcd
1051985 F20110114_AADKTP cho_j_Page_009.jp2
85b922223e9e4eccbfd5d14ab94a39bc
e2bc62b2f26daef21ee2b99d12d06156201dfe77
21548 F20110114_AADKUE cho_j_Page_064thm.jpg
8c679eade9d08861d5c0061de2001e2b
c4c5db61c0c83b8bdf0253a75a7bb4a4da9cd008
25917 F20110114_AADKTQ cho_j_Page_058thm.jpg
353558c3e72b401b30d6646b69bdca8e
c53c853d766015c8f6cf3024698de175cec4af90
195979 F20110114_AADKUF cho_j_Page_083.jpg
936adecda1565d33622cf552a3c75d71
8fdf2fa5201db20429e08da7dfd27680fdb6b2fe
29865 F20110114_AADKTR cho_j_Page_048thm.jpg
3fd55b9cc226e94397091c20f9f8c25d
66f85e50d8d89ad02d2cfc79f6dbafefaef9e8b9
23798 F20110114_AADLAA cho_j_Page_108thm.jpg
d307350c2498863f6d4211c5f77f566f
62740981ae222dbb3c82c73420be3a830157a0c2
F20110114_AADKUG cho_j_Page_106.txt
63d321fa825c5881e6d9ec10260d1812
de249991713a50efe4792b99ef54c237498b486b
82178 F20110114_AADKTS cho_j_Page_128.QC.jpg
7fd61e8538886a1913c4090f1780e589
6a89f7aab058a6bdd66c2c994aa106ffd4e9de5f
39238 F20110114_AADKUH cho_j_Page_111thm.jpg
d8b75960ec08cbb94b9d36cb53c9fae6
99baff10820f658c7c23cf67d0a2a5444cea26e2
84372 F20110114_AADKTT cho_j_Page_004.jp2
e6dafe766e6759d0b46c401072e4c552
673d5fefcb11163ec11f4794bfcc26b16399dae8
58583 F20110114_AADLAB cho_j_Page_025.QC.jpg
61aef043b22b928a29c63b97a9519358
04717926e0c16af443e49c52d822a7d54cb97618
F20110114_AADKUI cho_j_Page_108.tif
9b4fafc592d72c086bf2596bca6bbbc5
282fd460f43501cafc642c444721532e05e50200
46961 F20110114_AADKTU cho_j_Page_073thm.jpg
01b7715c6a4ceeabc8408af2592fafaf
68f68c6d4431bcc8a7ca5b63a8681b27ffe6e2ff
3181 F20110114_AADLAC cho_j_Page_002thm.jpg
cb4dd6f83eae454c32436ef4b745c379
9fc9ffc2129f89132d2dadd2bfb85e703056ba51
2373 F20110114_AADKUJ cho_j_Page_124.txt
47416ce20ec5da28543f64be8548ecf1
0f50e7c6ea22fae367fbe949908ddbfa022ab9ef
42416 F20110114_AADLAD cho_j_Page_013.pro
d3a8571d9db38387da75abc145eb8b80
55383f0b0df8bed6a6fabe6f46e766533fa2fd1f
F20110114_AADKUK cho_j_Page_062.tif
3663ae61a85cd5896a700a2ad82c1f9b
f8ff80bb439700e0961e3db7ae6e0341370929c5
F20110114_AADKTV cho_j_Page_063.tif
5ddd9d415b8df5e31c2d22659aa17a53
b9470e0075ec5dcada4341344fd24ccfd69ea030
1012 F20110114_AADLAE cho_j_Page_039.txt
34d606ddab7e6b25da24e3cdd3281677
8d9f2d45b62486b1be80fd0200898089a6155ba8
173624 F20110114_AADKUL cho_j_Page_100.jpg
74ad77b99f49adf843e814f99c1148ca
2d15ea2e943296c33f0594f782fba6cc46d75160
1875 F20110114_AADKTW cho_j_Page_086.txt
568ab12485321c29534f03d55a0b05bc
531750f1c6a3c1fc3efb9acabfa0d6c786e67ba1
25443 F20110114_AADLAF cho_j_Page_039thm.jpg
23e36b9c06df6812876f4e5e5bc33de6
f8ac0df1ce8f8618e3026ff883b5ff1b5e1dbfd0
254307 F20110114_AADKUM cho_j_Page_126.jpg
a23609810efe3079a084f2908e644abb
ad969fdde18d9feb0c3fab12ea91972d51957f73
F20110114_AADKTX cho_j_Page_008.tif
6f77c0c13f358d057c33ff7a7ab7fe25
24af085da59bb9b7d6d85f52642bb5c6f8d7e2d0
148607 F20110114_AADLAG UFE0008180_00001.mets FULL
f2002e917df36c4a0692cf1596a3167a
64a14257538809f0fdd9084b48d39c1256d20ae8
47876 F20110114_AADKVA cho_j_Page_106thm.jpg
e94370d5cc752c946b2210f7987e1de0
99d3772cc8e9f02451256aaf7767b93bc03da438
81286 F20110114_AADKUN cho_j_Page_012.jpg
fe2fe512b024979e2684a89ca5ce3f32
da45e6f33016a1ff1f6767cb0bbeb51beb664549
F20110114_AADKTY cho_j_Page_030.tif
d9e0028e8ba9851454c83f675cc2c33e
45aefc19f53258bbca849970a609166e638b74de
F20110114_AADKVB cho_j_Page_111.tif
2945fa1979662bf71cab886d56753144
87efab0a61cc028e74c09534ed97dbc3a2ec5007
245703 F20110114_AADKUO cho_j_Page_124.jpg
10f4c4669792acd1de3a83d3469a9356
b159c7fa11c1c3a5ea45d1de1f4ae7dd31ff318d
1203 F20110114_AADKTZ cho_j_Page_083.txt
cada827a0febffc448abc3e5a9a37c46
2125567f5442f3a1d9328972ec4c6d3a5292550a
205301 F20110114_AADKVC cho_j_Page_094.jpg
6a7660fe1e96687dbea06a601f046e56
48d9893d294897b87f35e3d5ab99585f83f382ee
71848 F20110114_AADKUP cho_j_Page_024.QC.jpg
425d933ca7c73779afdc1ddecf69196a
66a35de969439b72b7a00545484813f04fb1d14f
279981 F20110114_AADLAJ cho_j_Page_006.jpg
21b8cea4f262551355ff5d3bc8b68d78
e414c9193221791379b93114d37e8c8e034a1725
2472 F20110114_AADKVD cho_j_Page_127.txt
438538ad14c55a474a1e58c81dac0806
e99ee14747a3a2f10c12a8ec27c7c633ae152c5c
44322 F20110114_AADKUQ cho_j_Page_119.jp2
e297441de2d40ce4799527769dbe861c
e4f22607013665e9ac4dc60de16c0190aad2daf7
187585 F20110114_AADLAK cho_j_Page_013.jpg
8d15f10f880494b4dcf67490f1906b69
a456757ffe7d9e3d06662554cb16f969accb7136
23898 F20110114_AADKVE cho_j_Page_023thm.jpg
67b029c572d9684007541deb2718fe22
f78f6f5f540a4b14e8b7b84ff732591cc9c3cb8d
25692 F20110114_AADKUR cho_j_Page_127thm.jpg
f00ed463d94b78d0264578cf7c422e20
5a67233ed4b76c54d821ab2072128a5fe0bf1262
213554 F20110114_AADLBA cho_j_Page_055.jpg
3353c8431f80e8b43db0db0996c699d8
34caf3d750bd217993d942db34981426e7a2e615
225167 F20110114_AADLAL cho_j_Page_014.jpg
66583a793e5bfbada37a1be3ed494022
db8525b5ae368e3c2843f29fbedbbd8ad6b35761
180677 F20110114_AADKVF cho_j_Page_079.jpg
d7d04b3541fa5b8cc113a5cae8e39925
d4475b55062dd260b2d53b1b8689598d0bcbf5a4
85156 F20110114_AADKUS cho_j_Page_126.QC.jpg
498d3beaf7c4afb9f27d58772d49a58e
7f300deb480daf8e52f951a22de981ed9b176ca8
132359 F20110114_AADLBB cho_j_Page_056.jpg
e1e937eae0b93f142f93c05182e97981
1811af7ec2c428ce338059bbdeb3d1aaa55032f2
224581 F20110114_AADLAM cho_j_Page_015.jpg
54efd64161d75041172ec58559037684
c7588f83236d700f618ecad97d2ad3c9a1ba6ca6
126016 F20110114_AADKVG cho_j_Page_123.jp2
aa04640aacaab306c5bcb503e6347841
e5e6c2aefe254fb124fadfc026f752c47e92af2e
25275 F20110114_AADKUT cho_j_Page_031thm.jpg
c25b656908590864f4582946be3176f3
aad465a1486ec383f3a45f0a77a8bba77a8a1963
218481 F20110114_AADLAN cho_j_Page_018.jpg
4d9201251c995c0bb260f32ee1b169cc
fd7aea3c5fd286dc3ba9717721e0cd6963897680
1051981 F20110114_AADKVH cho_j_Page_007.jp2
d61f935c0c8914e2d42cca9036c79c66
58bdbf2fb0ef0b9c2448c35593897c02ed43b388
F20110114_AADKUU cho_j_Page_066.tif
a3d303ceb1ef1982a1830926b23c5d14
775c580a1af0063464a738900b527fc6fbf4578e
171035 F20110114_AADLBC cho_j_Page_062.jpg
1a2279cbc40d360ea3124aa310d696d2
bc707ebd10954016b3f599db8eeae1684c0234d7
149384 F20110114_AADLAO cho_j_Page_022.jpg
5c0e120ea1b7d0669bddb9ca7f0c238e
72fcdfc54757b873037ecba516d4a023a2b87e59
81774 F20110114_AADKVI cho_j_Page_098.jp2
e94b7a1fa4e22d3f1116be02f468d4f1
907df48bc7a8730c99da24d066f31a992cd729db
149157 F20110114_AADKUV cho_j_Page_052.jpg
dd850853600d2cf75b4a7a17b7990766
07ec06c98d55b735dea8733ab6acef9c066101f8
112640 F20110114_AADLBD cho_j_Page_065.jpg
3dffd5108f8836c7c21eb72c885b9b21
2f90e9169e6f318264a8a00325debabc7b063158
172395 F20110114_AADLAP cho_j_Page_027.jpg
8a72560dcd2a50f219902a4a8de2fb94
d82990d8931b5484eec8370f662ac08176a27224
75731 F20110114_AADKVJ cho_j_Page_049.QC.jpg
fd31086ee096be11d94f33a3dc052853
7f274374304c1ef16d09251c951f3bf1baac7c7e
164244 F20110114_AADLBE cho_j_Page_067.jpg
c194ac6c853319fe0bbcc89519aa98c0
bcc08761d5f71738ba1af49cceda12a578572229
171142 F20110114_AADLAQ cho_j_Page_028.jpg
8ded7c9880c36890bc44caf3dbf7b185
9335835384a1683ec7cc32ce78611a0371709f50
156656 F20110114_AADKVK cho_j_Page_021.jpg
63e44e4240042663ae2e8b50ee60d2b6
d4de51b937bc2b2699500aecd913e95721e1ba6f
86649 F20110114_AADKUW cho_j_Page_091.QC.jpg
77e1b3c9c4cc1bd6ca9a7e7eae95c854
9b0093db3fd14f441256a7e2133b804792f06e05
189293 F20110114_AADLBF cho_j_Page_070.jpg
df0de3cc4a7af12453b1a313c48b6f8a
c622f9daa103b283970d80398d0047c75240835a
184752 F20110114_AADLAR cho_j_Page_030.jpg
2daa02a9fb65f9d529a525eebd54a956
2f4ce4f1bf576f1ac5a1781f6ae13c21f571824d
29887 F20110114_AADKVL cho_j_Page_046.pro
7343d019d7aa3a2e5c82e816c2070f44
1a7a806c14cb9ce7a382247b71d9dcfb2222b1fa
F20110114_AADKUX cho_j_Page_092.tif
41a5d8ccbed56dfe68860608821c937a
de835d0c121d1bf88c87784937bc6ca9c9b54f42
180309 F20110114_AADLBG cho_j_Page_072.jpg
af4af5dedcc5f4223085203f976dc60d
91507a885e9d706a96329ca32a03c859c6ca8f83
37847 F20110114_AADKWA cho_j_Page_095.pro
ca54607966b4dc7073a518eb2420e4ea
8d873f7853f2ea06807d9d41c31d2f630caee15a
180216 F20110114_AADLAS cho_j_Page_033.jpg
a9aa924bd2ed813bb1992ee31c244c6d
3c1d37cb07cd7b8752f82b2193bfa611fe6595d2
178185 F20110114_AADKVM cho_j_Page_046.jpg
986d1f574bef18a39a23643ecb2d3c47
4fc32c7a644c9c518299433d6d5dadf4d25bda3f
43611 F20110114_AADKUY cho_j_Page_035.pro
a515dab62802e8cef6791d9db1a12209
652a8398aa2bfae9a17f7b3eb68120617d4dd74f
178741 F20110114_AADLBH cho_j_Page_074.jpg
76ed4e6aed32c8cb98c864f1b4182fed
33551ab33c62c433bd48968fec028530c7df5cfd
46590 F20110114_AADKWB cho_j_Page_017.pro
cdf04bbedff60400e487e9eac46d7ad5
de4f9b778e260d73e9d026d7c82c51bb4a53309e
211971 F20110114_AADLAT cho_j_Page_037.jpg
a87689432aa878c955b406ae3e9c6af0
bcd3bd2e0c2bcbd3fe8a9e1a6616e09264fc4424
105 F20110114_AADKVN cho_j_Page_002.txt
e562dfb5518ab066998780eaa39e4ffd
6b93c2a0c964e47d5c08401beeb1f44f95f89ae6
108758 F20110114_AADKUZ cho_j_Page_055.jp2
b39564a35dd52c55b0452f6b6b5ee2d6
a720576b3588f56e17aa7bd473407722436a435d
213011 F20110114_AADLBI cho_j_Page_077.jpg
a5723f3a7b9ffcfcc97717b32e263d01
672281a35a77a5200357ef002c97eb7ff84c1baf
1586 F20110114_AADKWC cho_j_Page_084.txt
05b794a7d56f476b2d7f1013e6ae796c
3e656fb922e22850e93af7b7233fc6c3e29e86fd
197458 F20110114_AADLAU cho_j_Page_038.jpg
d47dc569a8b0c8ec43378d8e584898fb
f99d9ff52f72222b385bea117fe4bd39f4558edb
1831 F20110114_AADKVO cho_j_Page_090.txt
2c35c027d029472fefddc7db393c078f
38d916bfe6221acf21799f879d05ab64926c0921
221332 F20110114_AADLBJ cho_j_Page_080.jpg
3889f216a930041ce26afe0ba72b0970
4f87b724e667f16495662cd523b272382b8d1729
27763 F20110114_AADKWD cho_j_Page_051thm.jpg
9241408b8d7bf8384a5d82a77683e42f
466990786d5e2993adec506eccfe217356d06db2
160227 F20110114_AADLAV cho_j_Page_040.jpg
1586de619333bb92250bae989d94bbd7
2fd829f854c9cd9a43e6a65576c08a7e328621a8
67028 F20110114_AADKVP cho_j_Page_026.QC.jpg
83c352c25a904c0996a1de02152a39af
ab5711aff448ab4c69974f71148156f66fa6a370
190558 F20110114_AADLBK cho_j_Page_082.jpg
14b5c03b513b7e7ee156506ba2288383
05a237aeba0d72462930aefe82dd187bc374d97f
22157 F20110114_AADKWE cho_j_Page_063thm.jpg
408a14b5cd5fb25e30ddc91bdf691533
0a3ca0352921299fcdc2fe84db6d9cf1fee2abb8
206562 F20110114_AADLAW cho_j_Page_042.jpg
5925235c1e4561b61e5dca2e7920834e
5c95edcb42c108a872ffe9c3663f615d27e42640
89634 F20110114_AADKVQ cho_j_Page_066.jp2
2c5a1987734f2ae3f219e7c6605b4a1c
aa454afb5a5112a25f1ebae3bfff203225e15ad9
190092 F20110114_AADLBL cho_j_Page_088.jpg
174a3d7205230860024b85243ba01388
9fd4f944c1d62d4fb51dbd46d9a25d2beb353eb2
1732 F20110114_AADKWF cho_j_Page_107.txt
b1da9b8d64eb7c5d364eaed5ae4df88e
48724f3d68e7f7010937cbf787367f90867a9b61
170850 F20110114_AADLAX cho_j_Page_043.jpg
4c641713ff854b972e1681c8fb602999
1ea1bb7cc3b71dce801ec7ba1f510151202bac27
F20110114_AADKVR cho_j_Page_015.tif
570d3e7dcb5471a7cdd0d1636746aa90
9875616774f0e770da113b7b5b7413162444b944
42223 F20110114_AADLCA cho_j_Page_012.jp2
af0b65133ef76b5142f2bcdda07a680a
63939286dc97e2ba7a7b0cc3ff0d988becbedcc5
190241 F20110114_AADLBM cho_j_Page_090.jpg
92304a4304d3c5c5bb3652b6ba542aa5
e659f330168922f79b9dedabe171b88c872c439e
F20110114_AADKWG cho_j_Page_081.tif
68fc430266a42888cda765da2f4f91a3
5c7319cc3c3d56dc31859eaaa4d80d2b21beaf14
224537 F20110114_AADLAY cho_j_Page_044.jpg
7b8497cf95e045e684ec596e396c715d
edd98a891aa121b07cbd44b7603d13c129525c87
F20110114_AADKVS cho_j_Page_123.tif
a9de7269287bad056bf5d7a9939a5378
f23a0b41958687adad59ff484c4db62786cd365b
93877 F20110114_AADLCB cho_j_Page_013.jp2
99430f7c60253bca563d07669e2a1125
5d4f7e7dd4c51ae6e60b1286dc9d24aa71b81b94
191361 F20110114_AADLBN cho_j_Page_091.jpg
f8b8eb43c153dec17673a9f663302224
dacd763ce85697ad9d7f3b74bbafafa0c28cfa51
1727 F20110114_AADKWH cho_j_Page_028.txt
6d1bc76a3a165249a0ae1d4bed4ef165
9908765466803ca2a74649047d8ba37a84bd1caf
205177 F20110114_AADLAZ cho_j_Page_045.jpg
052a7ee878ff9e76a15449087cc4c714
72ccdd122af6d41b36682a5e7f15822a6062ade4
1562 F20110114_AADKVT cho_j_Page_040.txt
493faf0efafb297a72ac045d2aa78e70
0fdd811cf71041d2ec0d636f19f952410d903d46
109696 F20110114_AADLCC cho_j_Page_018.jp2
7cb2fd0b4b86a5e8a7495e573cd95055
c61b4423e49e0bd14b7be64bf2fedc587b4924de
191108 F20110114_AADLBO cho_j_Page_096.jpg
61b41ec30b0fa7d2d979d3b05ea1223c
a6d50949a3a7bf27bbc077fd9a177d66ee6b799b
94953 F20110114_AADKWI cho_j_Page_005.QC.jpg
a33b13434cf7ce52a32e38b81e36801d
311d811f94b6818b8d382448472cf1de9609fd16
27127 F20110114_AADKVU cho_j_Page_065.pro
1fb2ee3b7b180f76a9d659b4701dcc0d
04ba6af439199c7f33f9056ef5f174b9320fbc73
185899 F20110114_AADLBP cho_j_Page_101.jpg
18418e50e6d12dc901ab60bb387e8cb1
286ead9bd33a59b6d6d85f0fccbfd4daf79b3b63
25635 F20110114_AADKWJ cho_j_Page_022thm.jpg
1bbb15f8e9ed55bf43fd2aeef5bff33f
a78588a09b2288c413b6b694e1737f491016b891
F20110114_AADKVV cho_j_Page_035.tif
e252b188e787c81843de7338114f5f18
73d0b523d0a575cc4fc47874dcf49122335cb7a4
754835 F20110114_AADLCD cho_j_Page_025.jp2
5d2e5624702a71133f81bb1fadd15275
5b8ad45eb6f1e04b1906d2f9455ca604d98f23a0
212473 F20110114_AADLBQ cho_j_Page_102.jpg
eae86f9f2b681d7408d9b443cddd9462
389c2b7e206d39835f8d416c3955cd8dfba42d69
46073 F20110114_AADKWK cho_j_Page_076thm.jpg
b785f19ae30974d23b6ebfd5d7c5182c
0ae075ce78b2a3e0db8233ac17204e759f69a216
166406 F20110114_AADKVW cho_j_Page_004.jpg
fc9381eadba6c412fce8e3bfe63d64d7
5d396285ec90b649d1a71bb5c3fe72067da2039f
809530 F20110114_AADLCE cho_j_Page_032.jp2
78dd31e633c50d781c4eb46c4cac466b
43ee8a50faa3f068024d6853bd4e7ba44b5f6c5f
183630 F20110114_AADLBR cho_j_Page_107.jpg
060e9bd30c1e603c520af7175e18ed96
e077b559776dfef79fe31324eb91c891c4fb97cf
973544 F20110114_AADKWL cho_j_Page_089.jp2
050254381b5560261722136955ef6d49
40e4132fbec17d2610f2ed65c02f6efa1b5d19b7
836553 F20110114_AADLCF cho_j_Page_034.jp2
0a99bde0bd3866746f24105f3851d6e0
60c798a2d13cabf01e3ad4b05c1ece3f44fc0374
66623 F20110114_AADKXA cho_j_Page_013.QC.jpg
30dc9e302f264a3c04a2338ee07f1732
824f8bfd2cda9b76a5bdfb7b463deb969d1877d3
208981 F20110114_AADLBS cho_j_Page_114.jpg
051b4f9fa402cf049799e6917d1fbc0e
39e5499f2da94aab8c11c0687a30d842811241d4
1759 F20110114_AADKWM cho_j_Page_050.txt
cc0b940ed0be7a11b466912a03a8cd5e
146d108167b787a053b4f9968a6e6238f254a2fb
1552 F20110114_AADKVX cho_j_Page_004.txt
d95babbf37e75af4cd2a6dcecd253050
56a02ff82a980ac7da6f801dbf1622253137f3c3
99374 F20110114_AADLCG cho_j_Page_038.jp2
99a5889d756442bb97f9026355e499fd
791b0b64eb46b0af091c8afe35a1708b6f809649
17168 F20110114_AADKXB cho_j_Page_056thm.jpg
59c52e30a260fe34d5dab1987807a600
a3d473233379f57ea5e4b8c6af9bb447a9f9bd34
47946 F20110114_AADLBT cho_j_Page_117.jpg
6945f687c500c360224a556b18bbbeef
c0e1cba7c760c429a809641d932266e9ad870ca8
95304 F20110114_AADKWN cho_j_Page_078.QC.jpg
61b74846f8b14d3bb61d4df4818c054f
ff0a104dc5090827b721f2ac7785d234e4da7043
776170 F20110114_AADKVY cho_j_Page_027.jp2
b027511e582ae2886d418e5a8f287156
e52a6f00949c2a4980e07b957153ba3199a3af06
98013 F20110114_AADLCH cho_j_Page_041.jp2
6f0787d8b26f77b50e978831919da2f5
483a5c8b455d31332b093c2685dddec8ff88985c
329292 F20110114_AADKXC cho_j_Page_010.jpg
7521066b0523ba537b4915ac7143e98b
03dac5b64db58cb092a62aed2e41eb6a68a6aa88
194909 F20110114_AADLBU cho_j_Page_121.jpg
6490695f83574a6a3a388f21641c0681
e5a8c1a4a4ef1625b9fe1cae85f54c3eaad007d2
13097 F20110114_AADKWO cho_j_Page_118thm.jpg
f9b7e6cccf35353d2845410f69d01f61
af2851d255e52d1ade09b157d89bba0726675812
219208 F20110114_AADKVZ cho_j_Page_086.jpg
ad08c2b9272400702b9eac9dacf3e37f
f9ee9f2b1ba00a3a230ad12d846764daa2a68655
848704 F20110114_AADLCI cho_j_Page_043.jp2
2d12df62f857acb1e426416246acc5f5
397133a2dcfd89404a35018e7a2d20ddc25d9d83
1015593 F20110114_AADKXD cho_j_Page_106.jp2
42bac7a39ab50ed3fcf4756beed8ab31
3750cb1b6857aac1c21ddc4829c72cd08e9ab112
242827 F20110114_AADLBV cho_j_Page_123.jpg
c21375ee17630ef79162abfe4685f337
a26883e004029afbea5352914520434d3ed9e8c3
71554 F20110114_AADKWP cho_j_Page_054.QC.jpg
a87d949d72db297c8241031e7d086f4b
784a9082d0bac659951b0ededd439444c2859e43
840163 F20110114_AADLCJ cho_j_Page_046.jp2
e88a140133649bd18def2a7dcba4307c
cdde479531ed06af189a86b57c3000b67d9c352b
1485 F20110114_AADKXE cho_j_Page_056.txt
d4e637df7337370e8fd44164337e9478
d5ca6b264b9a52aa7ec9b16c1f5eed2821b1d0f1
270953 F20110114_AADLBW cho_j_Page_127.jpg
0c91062f91fb3b5bc3ef3940ab86eada
044b2bf48d434d648edbbe8524fa23f9abee1ca1
202814 F20110114_AADKWQ cho_j_Page_017.jpg
a6cfd06557861d3c51c830f71befb026
e9de236c31ec346ca519bd432d3608c3614d76ce
966928 F20110114_AADLCK cho_j_Page_047.jp2
81c2c760a2144051076a6521274325b2
3bd57e3e986fadba644e5827c12f0d43430f3335
164134 F20110114_AADKXF cho_j_Page_109.jpg
ea1b4ac3e662d8a88004d50baca6eafe
e9c455c9b92c5ab918feec5bf6e87011cc5c05e0
251128 F20110114_AADLBX cho_j_Page_128.jpg
492789a5055b65c7bdfcf729ca6d77ec
71502b4aa617561d4af4a407a926ac50e77b1c6d
1051979 F20110114_AADKWR cho_j_Page_080.jp2
dcc6c899de3d4b368b5f1aa583a16177
fa7c6b047c8e1f5678612103ba038ceb2777404d
1051977 F20110114_AADLDA cho_j_Page_086.jp2
92f465d52ad4800e9c1f86d1de0ebebb
76a9eafff7c576b27ebe8ecf027998f72884377e
937660 F20110114_AADLCL cho_j_Page_050.jp2
59966b9b8817c2acb71f05dd6ddd66bf
48c0456f4087997e2d7fc7e76ad97e2a27bf6cf5
1718 F20110114_AADKXG cho_j_Page_029.txt
d7a04747f99d090f43e79bb0a7b24ff6
4a7b15939a26f3c609440de3ea2d16734583ae91
10641 F20110114_AADLBY cho_j_Page_003.jp2
b95653289189bfe1987dfd00312d9928
b11da7cc5c33bfb7e8fdfde44e24fe346cae9bbc
F20110114_AADKWS cho_j_Page_019.tif
ed12f263962dfcb1b2b5df7ae10cc3e4
d6f0f76e35cfc2b6c3b38b6d6748805b87b5b2ff
1034368 F20110114_AADLDB cho_j_Page_087.jp2
73c6b7994f1d67cf99dc969c6b9b72bd
10b7f4b56b6ccfccff94f206e1eeac08e3ad92db
767569 F20110114_AADLCM cho_j_Page_051.jp2
a9caf6b447a0fd1253916fdb211d5232
03d81309b03e58aca309d41258992add1dd26e15
84516 F20110114_AADKXH cho_j_Page_123.QC.jpg
c0cb9ead27039c0dbf62a714371fbeb4
6bf9fc9c2f8ce4640f614f9fce934f0e47c74fb2
79454 F20110114_AADLBZ cho_j_Page_011.jp2
201ef6c1cd9286f4d029484e10e20225
35f84807508f81b54c354f6d4c4357b3cd81b0e6
F20110114_AADKWT cho_j_Page_011.tif
f00f2c70c9a023491d8696f45f8e705d
3b932f5a1b456f26505b26a3dae21a220e8c73ea
1011497 F20110114_AADLDC cho_j_Page_090.jp2
8a03f76529ebb66cff824d593ef50c7e
5aaccdb978b554f4df7608bfdadf552723a2ae80
95674 F20110114_AADLCN cho_j_Page_054.jp2
d5e31a460ac0a455c1ce4c0cd0970c4d
da7601636229cec2c933b7afbc8c460572ba5854
32455 F20110114_AADKXI cho_j_Page_051.pro
e4550ef9a67014dea4e59b99de47c7f4
3533402c0e8137f6b78934e1a7ca01f82ac063eb
21203 F20110114_AADKWU cho_j_Page_004thm.jpg
809064f893f89bd25b24585c0b6e9578
8ebb124e4650e436ed2af385e62937c38ff4fb7a
98882 F20110114_AADLDD cho_j_Page_092.jp2
cbe6cc34cf3e9c08d6b0990c580b7b23
2686ee3f6c804a6b1b978b1c71db7d5031ddfefa
79516 F20110114_AADLCO cho_j_Page_057.jp2
9d7f109cf8a076b5d7c46e692ff74021
6bf495358530520464560ccb419ab79d296c4171
65283 F20110114_AADKXJ cho_j_Page_121.QC.jpg
602948e8865602e04a346bfd5db9c19d
96479f87019017f6f6dc368585e9fd42728e8b0a
28906 F20110114_AADKWV cho_j_Page_061thm.jpg
22e8a1c3ff7ade25bfff93c55ed3b999
0bef609987cd1a119a6ac03da380426783f42536
66981 F20110114_AADLCP cho_j_Page_059.jp2
2e81b021063100afda71c3b82ba671f6
f436d163847dc9df20f6c609ee11240d73f5c45c
10414 F20110114_AADKXK cho_j_Page_019thm.jpg
f86ad6a0a69fd11242b0c25303d82ddf
78fad1c4dab876b761b0ddaa980577a2cc12ce4a
22558 F20110114_AADKWW cho_j_Page_035thm.jpg
c26505c04ef049eda1cd56f7b8ce7c5d
a1535f5b8b887f77368ea01dad49316985991d02
577184 F20110114_AADLDE cho_j_Page_093.jp2
e2d054a0eb417daf2603009ead7dfdef
7125ad4a1649f8460ae662d4f79dff70f016a95b
87264 F20110114_AADLCQ cho_j_Page_062.jp2
1305ebb51bc507d5c4181b1ec1af1780
a98ae79fd29ac71f389790e68a706ef38a183811
F20110114_AADKXL cho_j_Page_043.tif
527cafd49947512c1a41d353575cde94
b23716adb75eb5167f7c168fb4f5039056d98d3f
174731 F20110114_AADKWX cho_j_Page_066.jpg
f5fa2cf5f965b8317260dbd1c3718456
9bd7edb80016c91a10396711f6cea17e363306f3
84123 F20110114_AADLDF cho_j_Page_095.jp2
9b0d0ae1d079ef372c88a4557cf769d4
55356029c4c35594f220b142a53ac834a919a375
86297 F20110114_AADLCR cho_j_Page_064.jp2
8ab0f716dc547bd5f241a0f6801d5d0d
e596e59e04100c1213e639387bcfba135a5fcb66
2003 F20110114_AADKXM cho_j_Page_082.txt
73ceef1288e2525b263f166479dbf48b
c7f93cb1aa04db0867f38701670c7c4765c89938
926321 F20110114_AADLDG cho_j_Page_097.jp2
fef0eb5ef0ac847c8641a2a3bc132927
ed3e2623929d62c7e04f702bf0342d5834da4cc6
F20110114_AADKYA cho_j_Page_116.tif
3db66ca5133c80b856fcd1d9edea4c88
94ffc8122441f3856c217bfaa6bf7cda29823a78
57644 F20110114_AADLCS cho_j_Page_065.jp2
0fab91448ea923ef20f715f5e7126d85
40047ec21c214e31f8dfbce4399919c056a75b0e
78178 F20110114_AADKXN cho_j_Page_019.jpg
1da66cf92bd29fbbc635d4d4a9008267
8b71306280f0070678a9258e19c24f0f510c9117
1034945 F20110114_AADKWY cho_j_Page_081.jp2
66712971f172ccf6e0e1d6c947081c48
ca0ca5f7ba833f63fa09590d92c923cc4a1a1aa7
887971 F20110114_AADLDH cho_j_Page_099.jp2
6c28886ef2abec861f9a81fe9a670921
0937d8bbdd205f773e604f3f121ea7a8802dd453
1969 F20110114_AADKYB cho_j_Page_031.txt
f0fa556c11efebb8afc712524a29ca1c
5ae1571ef70393747ea3e223f81b5b406b3e6c2e
75602 F20110114_AADLCT cho_j_Page_068.jp2
6c8345322ffcaf09a10a5acabe0e54fc
04941001bf59a43d9eefcdeaf0637834fa1498cb
93824 F20110114_AADKXO cho_j_Page_007.QC.jpg
8a42836f37ece4180e669aab5f67f3a5
ad18bd6161bbaa1778c0866a9a555854a00b128f
65853 F20110114_AADKWZ cho_j_Page_122.pro
101aa2a4d5c93c7fcfe6eb2b8ac96bb6
aea3512e540c8ccedea6e599554dd609a9b488fc
865034 F20110114_AADLDI cho_j_Page_103.jp2
50942ada16245540adf8ae5cf39b6a74
d434d049d8bd2c64543cc515baaf12bb98559504
73853 F20110114_AADKYC cho_j_Page_017.QC.jpg
0e3e067d471a8a0df8af745ab0356119
446190c7f8d65feb7f484c781262509ee60b53f8
86229 F20110114_AADLCU cho_j_Page_069.jp2
ff1dee760ce582d704e1ac9fe2ea142a
3e55f22ba2db8951affa8bafc2cbb32afb873116
136582 F20110114_AADKXP cho_j_Page_122.jp2
ddce1e005f732dba545ffba7f27083ad
f6ab2611a76aedef4dce8903f35737af050683bf
94968 F20110114_AADLDJ cho_j_Page_105.jp2
60954cc828583cd4e0f57bd54b445b38
8c8479722f145b0fc674ceb52464552c1bc9855a
39066 F20110114_AADKYD cho_j_Page_075.pro
4492208858e1e7dfaeccfcc204be4c5d
e8f3a82ed66a303e2c45991796261f3c3c3b8055
1021622 F20110114_AADLCV cho_j_Page_073.jp2
b718fc6afc423b05d70f854216f069f0
a98f7681b5347f33e98d03f3e33c1e7a6ee0bfcd
74714 F20110114_AADKXQ cho_j_Page_038.QC.jpg
fc66542ad5a47d4f805512e0c16e2ab9
947e8d02f1df3176b9911e13cb9e42f9c2c1dcff
101198 F20110114_AADLDK cho_j_Page_108.jp2
50d438940e1d9c3e18b3d4f6c0d97fc7
10c51f5329cfca89495bb6ee9002134da0fa6e5f
F20110114_AADKYE cho_j_Page_104.tif
56baf8c412c47ac4b3c0bb39a78fef73
85bf36d3e996ba4805451c58d99da13608666771
871799 F20110114_AADLCW cho_j_Page_076.jp2
95d2d49c06fb275698275a7596ea2edc
d24fbbce1a20450409ef4f9c086f233f71b4611d
F20110114_AADKXR cho_j_Page_056.tif
8bba152d839a22cd436c54ce26bd3cee
a0730bb443bd0ff7b924122c691acc7882f4f424
F20110114_AADLEA cho_j_Page_047.tif
35e76e54b1327661dfead150b1567fc1
2de13b8cb5756a645e83cad7db98d404df00624c
106509 F20110114_AADLDL cho_j_Page_114.jp2
aedff0568a5095369814a9067b613130
b752d2b86a806b0cc50c831a1927e8fcbf4afa3a
171740 F20110114_AADKYF cho_j_Page_029.jpg
13cb3b507ab07f0182f0acf612e64f18
fdfffd5a7198936734c16446938c15718db48519
F20110114_AADLCX cho_j_Page_078.jp2
92f3068f4fbe62a5645cb2ee31e029f8
2f66537f617b6d164162902e61e1319ee779f1d9
25141 F20110114_AADKXS cho_j_Page_003.jpg
dfec8ed4afc9a8d7a5f79fd2827617fc
8b8a87d9e5dcefbdc2f22a8ea79750ab9891d4d4
F20110114_AADLEB cho_j_Page_049.tif
5a41d74ecefa96d1cb10019e4cb4e7d2
e6963bac5d42e012af3bbd56b9a2746e82e8883a
114101 F20110114_AADLDM cho_j_Page_115.jp2
150da6e560d4b2f1907aa34e20b5edc4
a88dde05595c97bb8306f46512a0a70993df9500
22812 F20110114_AADKYG cho_j_Page_105thm.jpg
cf5faac07bfeba17cf4676be395a0da0
06153c8647c71f876d0d141a5e32578549db05ba
925491 F20110114_AADLCY cho_j_Page_082.jp2
b15a22c3fc1ca1adddaa9b22ac95cc71
84049470bc0493c9acee4b32c3e89c95425a3a2a
2517 F20110114_AADKXT cho_j_Page_125.txt
818ff3d6ff54b4867983c7e482da82a4
2041adf8e7fb6ebe33f77e34582fdacfe988ec73
F20110114_AADLEC cho_j_Page_051.tif
93890570a4b46f81cfed9dc7eff196ea
c9efef12dd84f9a9aeca5d30ca73c9e0c61abf2e
43358 F20110114_AADLDN cho_j_Page_118.jp2
1faa485681134096f38675f73ea15352
4f5fc0cdde647513aa6bbea0a0c4b58979fc8788
2004 F20110114_AADKYH cho_j_Page_018.txt
b72a0d7e4260b5636109c71fbdb64aad
a34bff714fbc5cdf4a37508c55a9cf5c95964de6
786714 F20110114_AADLCZ cho_j_Page_084.jp2
2fbb1a07431994aab5650853aec11356
f83c396ab2624e1724c10a41d191b40751dcb761
1900 F20110114_AADKXU cho_j_Page_114.txt
8e3087b823c2f605b617d36c065308d6
4c8a32549dba5e1ad36063480ceaa361623996c9
F20110114_AADLED cho_j_Page_055.tif
757d9bf2e5df83f566c524fbb82d3dec
6b7044414a2b802ad91e2a95e8f7e91cb5c71199
123707 F20110114_AADLDO cho_j_Page_126.jp2
d6c68c8b09c5ef483c980d867424f07a
e320cb9b34b0455c2d3b2ba7cdf59a18cdce7f26
71295 F20110114_AADKYI cho_j_Page_108.QC.jpg
bd44b344436e61a88247656646b8735f
0e4bca8e3f75ecb0bcfb15761c680d7f5b7ed0e7
30996 F20110114_AADKXV cho_j_Page_097.pro
f9a63692ad0d456cdb0c40a54db9446e
70d6eae831134b23d86e4a614f9d550eccd41752
F20110114_AADLEE cho_j_Page_068.tif
9d6d1e0934dca20071cb9d23eeb67564
12e9f5d4a0993e16aa4b635c364ef3c90fe5ed61
59587 F20110114_AADLDP cho_j_Page_129.jp2
892fdfc7b278d247d0644795363008e7
dd0e1e45f460ccaa390cac06094c26349e688243
19631 F20110114_AADKYJ cho_j_Page_001.QC.jpg
05a79c85daad067c0b0bdc1a94a383fb
44d67cf3d7884c1ca4a2fb95dc7f592ebface6cf
44265 F20110114_AADKXW cho_j_Page_070.pro
4c1a49d8063eac7add0f1e24a39f3320
9f812ca9f818fdfe81c2a07f941b2d519dc8d07f
F20110114_AADLDQ cho_j_Page_001.tif
9d162389676d0a787ee3a84ef3af8003
1d4c2485b34437e46344aa8d068ffc401a539dbe
22318 F20110114_AADKYK cho_j_Page_054thm.jpg
ca4b07be61f2dc884ed513ccda736e3e
fc0283a05e47ada17eb7bc35ccfba00969adc725
86932 F20110114_AADKXX cho_j_Page_028.jp2
cc8e248d2a0b097bfc3db121d3e1dd33
da27d4f8c4e9d4ed17d91ef51ced0a47abf25e84
F20110114_AADLEF cho_j_Page_069.tif
774392fdc1bda88d263ee060482270b1
bdecb6d2b257ef64013c04268cd7bef844454394
F20110114_AADLDR cho_j_Page_002.tif
b5c345f2063d35527c214ecd2f800ac4
bb87faa79cd9069f6a15c8c5411d47dcec6f12c1
257671 F20110114_AADKYL cho_j_Page_005.jpg
b297c0420383134f0e6d127b3c0e454e
3842ec6555b74a1b2312c4e2b349eb8fcef7923a
F20110114_AADLEG cho_j_Page_070.tif
871fc419c726a42c80f9141fe85fafaa
d45f7d7dce7ee2a8131af33c71aba44113d034fe
28248 F20110114_AADKZA cho_j_Page_101.pro
3d0200ddb61f5ff4899ca45d00bc93a9
3075f8aed7e1a6ac16930b4198c605e5acb4ac94
F20110114_AADLDS cho_j_Page_003.tif
84f561532e8b717665a2a5d2862b7471
c35178c769c285726e8c6a16c8496d055f70062f
20887 F20110114_AADKYM cho_j_Page_057thm.jpg
ac62a32442de37659ac51a3efabce039
215f1faf9d41f64c7d850f4cc4cb3375d131538f
1610 F20110114_AADKXY cho_j_Page_079.txt
4c798fc2be31250d2eb1bd25d6986fd7
6cb6e54684b988a04aca720e6b320a0903191689
F20110114_AADLEH cho_j_Page_073.tif
9a5d6a09c72d2bd89028ba6ef6c15cc3
ff3f492c60e71d9dfee1e5e3ee3793b09dc4dce1
222944 F20110114_AADKZB cho_j_Page_016.jpg
c7f9afbfa8d1128d105c837873409619
27023440d83dda7b9988b02282f3e3b9828f3459
F20110114_AADLDT cho_j_Page_005.tif
a99dab41bb28acd381921824b7668c20
d59ced944a2e56743c5151f77a1a6baee66bca73
30097 F20110114_AADKYN cho_j_Page_096.pro
065e063b003f695fb94131062e9c9ac8
824e0c33cdf750159c033e7db9ee4ec626e5f2d3
F20110114_AADLEI cho_j_Page_075.tif
5d63597d66f948157b979ff78845b4ae
446abf62affa349a6fc88d4d38c5aca0c965bc54
1150 F20110114_AADKZC cho_j_Page_081.txt
15eb5afa172183dfbfd6e7740a143d03
148ca75d69efe174689e9278981a8f81a33b669c
F20110114_AADLDU cho_j_Page_006.tif
695ee8a3c77fb5215cbc84132a0b46ac
602a375576ba985cfdad4379094bffb5fc674fae
136184 F20110114_AADKYO cho_j_Page_111.jpg
20ce1be7c61de4ea9262bfe77a5f1749
8d85b40fbbbc53437af443b259ff9706a37a96a4
176813 F20110114_AADKXZ cho_j_Page_084.jpg
3002c28d338db31ab000edd767dfba19
7f9b297336b78fece0fcf5d35ac95acec8f3fc7b
F20110114_AADLEJ cho_j_Page_077.tif
843fe341738c470172daf1e2cde329bb
8308b4184c8560e0ee2813022734546eaae6882b
59928 F20110114_AADKZD cho_j_Page_021.QC.jpg
b760093e57397d676ea643b7f4990609
53cc65d4497a9bae1562e6ccbc45d8e4023b5413
F20110114_AADLDV cho_j_Page_013.tif
8c925625e12c1aa3ea4c4b760ec92801
76fbe1ae87bc03d192ccb306d7b95c4ee9f9c381
1906 F20110114_AADKYP cho_j_Page_042.txt
c84284819d769c10fc937ca7298e4552
07882cd2351fe6b7163467f4d9d2fb138195a09e
F20110114_AADLEK cho_j_Page_080.tif
2947dbcc5debea62f71a995fc3efa791
26fe294ac7d8e79bb3252375f39a03a525742c3b
199754 F20110114_AADKZE cho_j_Page_023.jpg
1e4e4ad7d6bdfc12fb0a097978aa018f
8bbf51fe70a1dd410c8b63117972194c8c28c815
F20110114_AADLDW cho_j_Page_018.tif
4818ee7bd2f1dd5089c8f77184838a88
61134fc260c97c7f5a45334da83395de602b48e4
46717 F20110114_AADKYQ cho_j_Page_006thm.jpg
9584e547423711a8bd0f9228ee497a73
948e000a2fea6da3d62e28a968fe483f11237be3
F20110114_AADLEL cho_j_Page_082.tif
db14c2c7c5ebb8ec19df81709eaa04ec
0c02b7a73728dbe7820bd729e38c87be28fa988f
22416 F20110114_AADKZF cho_j_Page_028thm.jpg
d21518c4b1a8ead8e00560815897dcd7
5c513be3041a47b42f6098a15f6d53bfecaf8094
F20110114_AADLDX cho_j_Page_024.tif
09442843df16141d04c283a3c2638fa9
156f6d8447f80a74614ea246d73208094f88a552
48258 F20110114_AADKYR cho_j_Page_077thm.jpg
999ce4e267446ece1dcaa0cda7015903
5fc756ac98c09c9859ae232006e411ddc7d16090
F20110114_AADLFA cho_j_Page_127.tif
2f26148137eacd611cb016c30f64769c
df70737e83468a131321f3cba850e6ac50a7af82
F20110114_AADLEM cho_j_Page_084.tif
34fbfd25c9059d813fde62468a783412
c73ee977b26f6e4fdacd217e130cd0b01cd1d173
47215 F20110114_AADKZG cho_j_Page_105.pro
abcc279fce70c5fd732a762c604f73ce
188a1f9cedb33a0a970af93364a33ead568896ca
F20110114_AADLDY cho_j_Page_033.tif
7f319732cc8c8006b4f11d24b43d2714
e59a0648e878a5be4aa65e55614d83b00bac3719
900494 F20110114_AADKYS cho_j_Page_075.jp2
b7252b8dec44812022f4773e777d04ee
f384b2ef728441912eeda68ac2b9dd2030e70ac9
F20110114_AADLFB cho_j_Page_128.tif
4b6c4eea2ee1b4ce60e8d96b8e42947e
2cc2c6f29b21c97f297d9488b51fd429491e3975
F20110114_AADLEN cho_j_Page_087.tif
5a85e2561eb8612ac8e7eb69c1d16261
a59ce0b4b8a973bbd3c5d719b94a462ac660630e
189962 F20110114_AADKZH cho_j_Page_054.jpg
f083f49d99e2c4482f16e9f455543f9e
afe097d7e6241a8fd2615908fffa36f213eee9f6
F20110114_AADLDZ cho_j_Page_042.tif
e0d4349e9ddaf1b99e3d332742169653
377a37bc3987507f0dcd488b2816ca589e2002cd
61501 F20110114_AADKYT cho_j_Page_095.QC.jpg
e14c30b972969cf931c09b97af584bf0
be870ca6d93971d0a7f3b0f6c58b8a56d506c544
F20110114_AADLFC cho_j_Page_129.tif
955158daab6df2bfe91cc78f503cd308
aec2a07f8102c16bad017def7dddf5a3c582eba0
F20110114_AADLEO cho_j_Page_094.tif
bcfbcf89cd9611f8ca7b7f729670d04a
2412e4b95845f7ef37e098c6f10b9d0b6af75090
80604 F20110114_AADKZI cho_j_Page_067.jp2
fb490769868b45646f77a5db835acaea
1342f6ebeac7b0d854707fa8d25070d1f2ca97e0
232115 F20110114_AADKYU cho_j_Page_078.jpg
c52355f8c139c3a8c5768ee39c57df79
bf10dc089342fe32e489e4ca5af25193e365c864
1068 F20110114_AADLFD cho_j_Page_002.pro
d4c191e8310b1e26c4574e60f24c8ff9
ec00a5fa2bb45d8f121f846934f2fbac2f349bc3
F20110114_AADLEP cho_j_Page_095.tif
73511478571299c8cafebc9d5bf9c49f
0afee433e92e56ad48f7e146572a6d8c368b7bc2
23087 F20110114_AADKZJ cho_j_Page_074thm.jpg
e310b03df507e24eaa67ef1911bc10fd
b48f20e61ab6b32303dd460690ed4d9ff20c8a66
41538 F20110114_AADKYV cho_j_Page_064.pro
7a831412295e586855c59aa688c4eba4
897fbc71ec6597d5c2973336547bc9a306311e16
3549 F20110114_AADLFE cho_j_Page_003.pro
b08ec2444aa9929c005385013c9d076b
edf3ddd0fae9638cc1ccca586beb38754cc33dc5
F20110114_AADLEQ cho_j_Page_098.tif
918d0cf95480c42674ed01a44cbe9186
c6987527c8495657af658d073abe8e6949a96081
69401 F20110114_AADKZK cho_j_Page_046.QC.jpg
46706aa97f53800fd878d5923b698828
181adacfd53541230db7da61e095bc105af852fb
40714 F20110114_AADKYW cho_j_Page_019.jp2
8b1199ec7fedee3903b73ebf3ff817ea
8cc7844c85245bba2aafc79c657536e95bdd289c
76887 F20110114_AADLFF cho_j_Page_005.pro
18b58921cc28c04e9e620e6bf58d899d
5683f8e48d04353a1a203c32737cc9fbff92b9c1
F20110114_AADLER cho_j_Page_100.tif
3573d30faa035bb453a7360873477f6d
d95d3a45abe4210d889655d5879955646fb7f592
18715 F20110114_AADKZL cho_j_Page_011thm.jpg
7a705084d35fe8557faabc05e970e636
de68eecbef0431506dcd81f7f534638a974c58d9
98986 F20110114_AADKYX cho_j_Page_086.QC.jpg
628d80589cac09996f2cdfd6f8dfd5db
03fe2930bdb94668575bd9eda09183bd3780b424
F20110114_AADLES cho_j_Page_101.tif
033b14815369bb6da1a824295e171abb
dbed8ad69318637f005b174257351d990f724407
1580 F20110114_AADKZM cho_j_Page_080.txt
6d7219aa82fce4960175f2dcc26e6c25
a7793e73e9475de4ab260221aed7f6080d724a73
100113 F20110114_AADKYY cho_j_Page_024.jp2
c4a5dcfbfca9009b6511d47d16400132
f1565898af812b32a58493a1fe0ec5c363f649f6
83539 F20110114_AADLFG cho_j_Page_006.pro
a1ff2ebc939df7644e0c40a250aa6c55
56e488287821041b22c1fc8f4f856d0b31c7858c
F20110114_AADLET cho_j_Page_103.tif
239b0a8ecd98f882f860ff05c293a887
9f3a1e7428d0766a29e53ac3a921db3cefc63847
195743 F20110114_AADKZN cho_j_Page_024.jpg
dd755606bced3a7730dde95ea7af697d
fb56c0e90089d4fdfd80c5ee117fe5682bb11bc4
1262 F20110114_AADKYZ cho_j_Page_046.txt
52f4516198346c7482eabd33a993324b
558cf29d30d99f91e954b24678f63fa96e3063c5
46635 F20110114_AADLFH cho_j_Page_007.pro
aff1675049830373e8b372426657ab04
1dd40d78264d16c5a137fe5f2fe39068adf91958
F20110114_AADLEU cho_j_Page_105.tif
d5b5e8b483fd7ebfbb0f108f96c3b5d9
bcc73331165af9c4a505b93e31d1f2c1cfb6b429
F20110114_AADKZO cho_j_Page_097.tif
3d75b31f73b0975d34b0d23a4cbbe7c5
80ff4d20e337ffe304b943a576ac00ef181d95f3
53006 F20110114_AADLFI cho_j_Page_014.pro
1f7177755cd2e018cbc2d070b7b4c084
ccacb527912327015521baf67bc179ef9d040fa8
106110 F20110114_AADKZP cho_j_Page_037.jp2
6a4cff4f9f43bbffcc715f56f0226d7d
3d909a0617384d756fb365d3cd61e62b5b94958e
28976 F20110114_AADLFJ cho_j_Page_022.pro
1ab8d1784913a29c39e8ca143bf2813a
08b8caea914e4a3b8f11515487418f71281cab46
F20110114_AADLEV cho_j_Page_109.tif
3db7b1c945ced1b27cad247648e33600
8504af41f36641573f69d4a7ea3cca23d4eeb2a5
F20110114_AADKZQ cho_j_Page_099.tif
f03b73c60d9305a44dc2f7c3c53ac52a
24196e1e8b969d9a7dfebf2a8b90b575fc4f66bf
47745 F20110114_AADLFK cho_j_Page_023.pro
fb4fcf7c53018c2ecef5abc95098aa50
22d7d35af7ddd2b8f60da68e216b5da02538b08d
F20110114_AADLEW cho_j_Page_110.tif
55bf1f182d65143c34c3bf1b781028e5
bd81199e858787963085a94ddebc7bf62cea5b80
24617 F20110114_AADKZR cho_j_Page_070thm.jpg
3027e43a444c4f720b11069919b78c41
7940bcc6121cbd07e40b93060eb3d947ffb3c331
36962 F20110114_AADLGA cho_j_Page_080.pro
cf8260c994baec8908eb93a2f1d37f68
a21ebbf5f39d2d9c46027e80783f36cb104a320c
44763 F20110114_AADLFL cho_j_Page_024.pro
93f19ad2ddc162f00d47a6546ad1552c
868341bbc458551447b529fa8bf4ed3449de42f7
F20110114_AADLEX cho_j_Page_119.tif
44b802c3b36bc7c846e734173adeab07
72a6392d56c20eea81ef8b837bd45964c8b6e435
24043 F20110114_AADKZS cho_j_Page_083.pro
f7d41997fe26aa046dbd153bd476beaf
e9cc584cfcc1913b3b4e40dbb8e0d5f5050416a6
29740 F20110114_AADLGB cho_j_Page_084.pro
2e97cfcb9986dea62eab7a6e150d237a
05b08df6c308c4f070110a11ccbd4f7a07031668
F20110114_AADKCK cho_j_Page_036.tif
2a07dac2c25e5f91ffa774226273f3e3
44ff0118687dabaa8b061e05f56e110c5718370d
37320 F20110114_AADLFM cho_j_Page_025.pro
c75237d6450dd80fce4c6203c5961f39
f51d70fdb4309f36bec4e8341e5f5acf4fb35f37
F20110114_AADLEY cho_j_Page_120.tif
e10c503bcdbc5a6f255a52ef95058594
8442ee117d68d623b87bb9ac8f94b2c29a8af973
F20110114_AADKZT cho_j_Page_074.tif
aa6901d2bec14a909ea09b222a0dc901
2d283f304df59b131d79ce69f004930c36e9ad46
84216 F20110114_AADKDA cho_j_Page_015.QC.jpg
488abb71fc9826de5191aace26ddbffc
ea425386150c986e5baf434a6e00b7c83b6cdc7d
41896 F20110114_AADLGC cho_j_Page_086.pro
9f8c930eab0acfa6a54e4b2fb23328a2
f29f0870eaf9465628afa736d2268bb7c50f9060
F20110114_AADKCL cho_j_Page_038.tif
4de9f63b6cef755f62e21d694fd70ddc
de4fdbe539d519a640252eb8f4faaf32d30ae6c8
30184 F20110114_AADLFN cho_j_Page_027.pro
d5b05522060136417e26cd446c7b8a81
d1f9a31d1cb9b9837151a79cded08ba5829e0645
F20110114_AADLEZ cho_j_Page_126.tif
6f8704ebe824738553dd61784b0f1817
3e79042f8a0e13888f24cf72067dc67cfd17365e
F20110114_AADKZU cho_j_Page_093.tif
73d06e3a0d3aebf71213121af73c9782
15b176ec2adbb8faa682f819b57b27154999438a
42025 F20110114_AADKDB cho_j_Page_066.pro
561424c194427af299ab9d6386cddbaf
dd674f942d5c17d3c862e18898c25d27196399b7
36341 F20110114_AADLGD cho_j_Page_090.pro
b5bfbd7c6a299ae5865fd63124c4bfcd
53487445abd23ba01b5d4acc06b8227db85b6723
F20110114_AADKCM cho_j_Page_025.tif
ecadc59a4141dbe425d8d96878a2c8df
18623fc519ab024a8d614c13fd0e20a3421631c5
39952 F20110114_AADLFO cho_j_Page_028.pro
a7f141fb61b474afebb2194a8434686a
bb715dc9bee5b4def24a8c9ba944ff6c8acde9f7
185609 F20110114_AADKZV cho_j_Page_061.jpg
1b90847ea07ab9da25c77d681a2d4e8a
569f6abc416ef033c17b31ed273aaeffa6042d44
22728 F20110114_AADKDC cho_j_Page_060thm.jpg
d273b87d53c24f217fc05220389c681d
bcbe167618d4d3ab294928dad3146db954e925ea
22439 F20110114_AADLGE cho_j_Page_093.pro
1f98a7e4b49f70ce740b8e75e5780dbb
0e3a82853e7f6ce7a750e68d7b0464f4e3effd1c
34969 F20110114_AADKCN cho_j_Page_091.pro
810e0aec74c838de9adf877902fb80c5
fb8a38495850d2aec7dc80af734ff2d3add3e808
39874 F20110114_AADLFP cho_j_Page_029.pro
df01a9165d61884d691257db319019f9
4bdb452d494a702116e7cd71be6a8dfea3d0788b
545435 F20110114_AADKZW cho_j_Page_110.jp2
cb3b5f8323962eb68a8e111b7eaa4552
d3c9e6518ee84af1edc0a2b6a7354b9ebb286e44
37777 F20110114_AADKDD cho_j_Page_004.pro
8ee6b3b542cdf840e4cd8d77b50ab4b4
d649f58c621a3023df5cc0a01252e85b16e43b43
30779 F20110114_AADLGF cho_j_Page_094.pro
f0a76381988270740ebbafc6d474239a
c7b3c8467f29f01946433eb8976acf92d009ce92
44359 F20110114_AADKCO cho_j_Page_071thm.jpg
5fe42f9bf99f9e834ee9be1a5bb7b66f
c43ec90cc71d85c239520db75776064121129844
34879 F20110114_AADLFQ cho_j_Page_032.pro
9f51885c95cb2d8b55f60116a2eb07a8
04feb79c7781729a702e7983380ce146495bd3fa
82657 F20110114_AADKZX cho_j_Page_109.jp2
8285633ba20311f74e2c6eea85cfab8e
a52de84d7495ba4b2ddbbf1e31da24c8f8655b2a
45898 F20110114_AADKDE cho_j_Page_038.pro
8f474d61f37226482aa8ddf942d4c19e
a3b644dff5d78545537a480dd48c5a3575cb0e2b
35440 F20110114_AADLGG cho_j_Page_099.pro
01f4e6bbaff44b064e8e9608545ed645
ace8bc0ea86c2983c8481793506fe38c66919833
850374 F20110114_AADKCP cho_j_Page_107.jp2
3853551722dfcc170db3a1d145874e2c
66f3c860a1e8ac997f80d6c12cb3c4b4c1576fd6
22128 F20110114_AADLFR cho_j_Page_039.pro
17405dd17fb1db09908d4a74c489620d
6a333fbb1b25a23cc1486cb2a8d35e9611685366
F20110114_AADKZY cho_j_Page_096.tif
75ed9c33547430b9843988400b7752d0
cc226d88441762f9d766589a23b60a413555f1cc
52972 F20110114_AADKCQ cho_j_Page_044.pro
7b6281737b91760d393ac2cf6a9e52e7
1517b2ddb5056546194cdf5e630c8c171429e2e2
38261 F20110114_AADLFS cho_j_Page_043.pro
5c1bade933dfc0e5b9652ba7100e63d5
1a63a3f4d8613b21df36923b73285fc2418eafce
F20110114_AADKZZ cho_j_Page_112.tif
b747141769601d86c5172a9986c0c1c9
12bc7b91261fd45f57a5de9c6e15d613acb55ba6
F20110114_AADKDF cho_j_Page_016.tif
196ac527e76da3db6465939219a52f52
84f11b26dcb12527cd577657922089998027895b
35421 F20110114_AADLGH cho_j_Page_107.pro
f0c6e1ca5739173c4763ff74b199c335
91a3b486f6ac533a91e5390d63af2b27fd733a8a
972999 F20110114_AADKCR cho_j_Page_104.jp2
593551e5a5290343ae0357a2dcf7622a
308730336e87b8f7e8a5e6d0cb699e498327b7d1
32732 F20110114_AADLFT cho_j_Page_052.pro
6a7183c8fbb0e6a1a6e478ab2b6ec918
4fad328d24acb8af02b7d87ce522e77c8c286304
706923 F20110114_AADKDG cho_j_Page_111.jp2
c63d7793dcfbe1492015c2bf79c0f1dc
6565d25f204b21a016859ea1ff9cd8956c02ef34
10154 F20110114_AADLGI cho_j_Page_110.pro
196f0c3159facec3ae8f0498b3953391
fdbad53183ba32dc6feab6a6cbd4f5ede2536420
93731 F20110114_AADKCS cho_j_Page_030.jp2
1b7c7da8712f7f8f8efe7a8458da04cb
f95aa7e5e7ebf8916ffef74db1b2c82bdc8a7f87
33269 F20110114_AADLFU cho_j_Page_053.pro
5cd6b678ccdf90f87dbf6b2d068bdd63
e45141bdc315ce79a3329c81a8d8cd90143e4a87
F20110114_AADKDH cho_j_Page_091.tif
bef87fd6805c743794d3de4f2df986e4
f44ecf1c45b1a4df377683c2ef2531123b83be34
10025 F20110114_AADLGJ cho_j_Page_112.pro
b0d58b05b617a909838834c186ae349b
02e0be02515acedf17e465b9a653da13c7ccfc5a
82924 F20110114_AADKCT cho_j_Page_027.QC.jpg
db3185c835b09bb0bfde5a605d99cdf4
321e262ee15da193049fc19266ce24a9839a75c5
43834 F20110114_AADLFV cho_j_Page_054.pro
271ada9ed7079d9c4db1a4d214bd9f82
f2c06feb1bb2f22319c4ae841734bf84a6b3a0b9
34760 F20110114_AADKDI cho_j_Page_076.pro
ab521f5ebd7d6510b2ec0f9a22f4c20b
cba888e753578e3e7682ffdb500df1f49056984e
43183 F20110114_AADLGK cho_j_Page_113.pro
92a80c293914e976e9b14f75689b6dd9
ceb7cacfc8764dbe7b5fab3d7cabd5462d6c1b6d
11324 F20110114_AADKCU cho_j_Page_111.pro
03dc41e37d2816a6365277e89b74214d
0e48aac5dd2d605f0316cf86ff33f1a14747f1b0
29821 F20110114_AADLFW cho_j_Page_059.pro
6258e36c89ebe1c8b1eecc4ca8973a9d
e22fad8857a9c81ee112ffdb02f5cc39202d2e14
1879 F20110114_AADLHA cho_j_Page_017.txt
db627bce5e42b6f3f5b3ec14c6280d5c
288dc4738ed5c7839cbbc4f20728a62ff7f92550
174720 F20110114_AADKDJ cho_j_Page_026.jpg
a3c4fdf7c26137ccc896a26ecf7c42f5
003f63771acec13356a53eaddc4fa37f1e3385da
48154 F20110114_AADLGL cho_j_Page_114.pro
b9aefc14ba4c4ae702a568406b570aaa
cf340cb4ef4722282861fb80269900602d0105dc
39368 F20110114_AADKCV cho_j_Page_048.pro
65aa816ba996d337d57e22f61890c640
c305dbfb128edb6dcbe5ca63d34b5c4968e373b6
39131 F20110114_AADLFX cho_j_Page_061.pro
aa97742a59630ca2c2eeb9ecb315ab54
718767781e231910978698b2dc8a7d42f8f98a7a
1827 F20110114_AADLHB cho_j_Page_021.txt
4e36a87c1743ecf57b945acc9245a759
e56048f6a4a5177ed7067fd0afaae28fb55308a8
1666 F20110114_AADKDK cho_j_Page_043.txt
3521358453de7c127e06350b3a4973f8
3f9535c175cab3e7c484d809629a9e9a3ff74123
53493 F20110114_AADLGM cho_j_Page_115.pro
07a0b5085a7a313521f25862d65aadd1
c496b3012c17e0d22660c2618b6abb0da0f2cad8
F20110114_AADKCW cho_j_Page_031.tif
a61a8a461ac97de6f903dfef35ffb150
7f725d38ea6a498870594465cdc4553bc6eb06d3
43766 F20110114_AADLFY cho_j_Page_077.pro
2e83415b22533d8464f292b70ee36d35
df201807bc5cda12ea52f8f0a26f6158b372f176
1836 F20110114_AADLHC cho_j_Page_024.txt
35aaea06de35800e8941f2355d7f0481
2a46e418e8960f787b426051a9ceb0c2cc382b55
160711 F20110114_AADKDL cho_j_Page_112.jpg
fe18741559ba8bb4c8f6cb859a991cdd
57147d270bd0b9cdf371cde5a789a51abffa3e8d
9539 F20110114_AADLGN cho_j_Page_117.pro
f30cae3bf795c5267504e2b55da3da6c
9a505f2cf9c5ac1bb604d78b4140e606831be510
48150 F20110114_AADKCX cho_j_Page_031.pro
bf9069ff4986879362038caba244aee2
d01fbd311794207cbbbd74934162f40fd3b96c04
32207 F20110114_AADLFZ cho_j_Page_079.pro
af2128378814d4655cdad22f8f541024
7e4c1ebc3fe4c40aa21f345696d39f3dd10ee83e
46670 F20110114_AADKEA cho_j_Page_081thm.jpg
4060483535a17cebd949d13d81c0cd31
2f8b9542a4fb8264347dd56149612773bcfbbb2f
1669 F20110114_AADLHD cho_j_Page_025.txt
0424b5e37202eb6e17c0de5c0d0602e7
0d47b01e25a162a0b695a44bb3f14ec3662c880b
103758 F20110114_AADKDM cho_j_Page_031.jp2
e074e1baa06dfb333edd22602452acb6
dcb7024b7f72dacba8865b0f13aef13c672aa34b
23635 F20110114_AADLGO cho_j_Page_118.pro
9cd463b46d988fe7a6f5a7dc62b1545b
5afb1c93c8c7de652d43d36731d02c054b7b716d
48966 F20110114_AADKCY cho_j_Page_049.pro
e29f04cfeefe2abc44459eac0f69371e
593796d2d8193b08ac0f2a36bdfe2d6248b5d0af
37664 F20110114_AADKEB cho_j_Page_057.pro
b95ac629e7f3422cb69c0731e7c7716d
92d1f6444fbde753ad9b2f5469b4f5afbe0eb01a
1521 F20110114_AADLHE cho_j_Page_032.txt
2176b3a5345d63347c2cc0fcaba5a5dd
9e1cf1d17923b4f87ef20c821c7c3b8701478392
96370 F20110114_AADKDN cho_j_Page_070.jp2
c4fe66318a8f0d7aaa2b92b534b4b3d9
bc3b80b04dd2475b2ccaf04797e3edbf88c81ff1
44079 F20110114_AADLGP cho_j_Page_121.pro
1a82900fbe758c0681a7bc077268a4cd
8bb07ba262c86075dbd7c78cb8499ef4f6cb341f
183779 F20110114_AADKCZ cho_j_Page_089.jpg
416b6b21d27e42104b81d84eae4cf0ea
c112ca21450b9cc64a302f431226f3e7ee5af881
19895 F20110114_AADKEC cho_j_Page_109thm.jpg
6c640b30dbf584b86cd0cf6acbcc0f61
2f26f295e04948a43d12eaa8b29d93bb3fbfcf57
1522 F20110114_AADLHF cho_j_Page_034.txt
01d0d2eefef8d173e28e9f4c28a84980
eea4e7f7f045f6e3b3c85e3654d7ab927f437c97
59808 F20110114_AADLGQ cho_j_Page_123.pro
fea9f034f1c2bff9676847082b01ca68
270422d25cc123cb3fa722862ca605157f3ff5b4
107308 F20110114_AADKDO cho_j_Page_036.QC.jpg
d267e52585d6038210bc6efaddd6f705
2c8da231fc444d728d9891fa88c3f87efd31308f
55107 F20110114_AADKED cho_j_Page_011.QC.jpg
cbbb9112435539665bef95d9a3d18892
426517ed3084fcffb840a1f32be34ad29bf95872
1911 F20110114_AADLHG cho_j_Page_035.txt
ea92c70fe17c5ca125f4ff2ac9df24aa
16c3f2e604b626ab48d962dbad58f724a73c54d5
61250 F20110114_AADLGR cho_j_Page_125.pro
c5f365dd2205e49edfe69eb48de8e729
e85b0c584e4bbe3f781e51149c79fc1c9fe9a73e
155289 F20110114_AADKDP cho_j_Page_068.jpg
20d75bda3b1eb9075ef379d2c6cf6f18
c721671f8e2f2b04987d7dc2a5bc57f97d022b1a
F20110114_AADKEE cho_j_Page_090.tif
3e3d8c3769f7c466f3a904ff903f8969
2c3f8a7a72a28d3f9d3b93d2156c02cbfaaac72c
1979 F20110114_AADLHH cho_j_Page_036.txt
3e9a8babd629e731ebccfd1c60a720ed
75f3c9d3c2172fd4b101199bf8db6841b1c1ab04
59658 F20110114_AADLGS cho_j_Page_127.pro
eef50e0700abe124f414d79cdb46daad
567f1f3a4fe2c6d91e31a62a3c4bc74065de39f7
159300 F20110114_AADKDQ cho_j_Page_098.jpg
94ae398ea06f93d90d89f58be7fe8f33
099c5661dcfe3727d3e3f0ff4ec8c0584511a1e1
F20110114_AADKEF cho_j_Page_004.tif
5dc92a7984740b0a0d9392fed37f3a7b
2e83fed4cbbcaded2152d4d2d3fbc0fb6340d20c
17512 F20110114_AADLGT cho_j_Page_130.pro
ebbef521c91fbf86dc4a734436c8318a
4e38937b3d57bb921b24d2b8e7ee00a4519ec240
139961 F20110114_AADKDR cho_j_Page_058.jpg
43fe34cb6479adf977cc5f13e8f7ac96
4ed2f6b8d298d22cf8daddf90eb6048210f7cca4
1950 F20110114_AADLHI cho_j_Page_037.txt
dcbd5e21fdd0da6fc274ba29d2f21ae5
a73ca711754ce6bd5272921441d9e96ec772380c
3398 F20110114_AADLGU cho_j_Page_006.txt
9f65914b02cf985790367cbf221dd5e2
62635553c1e69e7e99899d694b01d6e21ba0b300
24170 F20110114_AADKDS cho_j_Page_117.jp2
7a0cfa4242e3320c4c4fd213b6652c5e
9a86b38ee756ea9507f42980fda7c8c6002e7ee5
44466 F20110114_AADKEG cho_j_Page_065.QC.jpg
4947b808ad94af84916e065024ee6a98
0a4006d27e9fbb7f83f7a187f07bd00a9e469498
1859 F20110114_AADLHJ cho_j_Page_041.txt
ece394161b98b4e38d1ee006e671eb66
3e6cd34341ba98b37899f68ea4c61e639f61acca
1928 F20110114_AADLGV cho_j_Page_007.txt
17479a78cbe83c9ba79696ed18e0f6da
8c54f8a1012185b98c1e1216a8696e06a766f754
44444 F20110114_AADKDT cho_j_Page_030.pro
198bca3cf8e84b0d8ec7b210a4ef8a8f
90e140fb1f9fb3d6c990c6420622b9d6d741c7af
F20110114_AADKEH cho_j_Page_039.tif
6af1252942a0a17b5a718a80f001b384
a6fd544b18f9ff550b3c252f7f8b25060a9e27f7
2102 F20110114_AADLHK cho_j_Page_044.txt
bd49913d4dd3213d87d3a87d96b2f220
f1ab8ba0323483a5d1d550a63908156f0c98151c
2200 F20110114_AADLGW cho_j_Page_008.txt
47fb3ad68e4618255218aeb5b27016b8
98744865c5bb894f6b5c0cfa325c7f8eb32fd5a7
1486 F20110114_AADKDU cho_j_Page_109.txt
0fb463fd9171b17b7bbba05eaaffaa7b
784adb34b1f2dfe47d78ecde442a6fe0f6a60221
5560 F20110114_AADKEI cho_j_Page_002.jp2
4452da81b2e5f13062b1030d2c7addd1
8173d93d6cf4e852879d64f425bfa0d69dc863f6
1651 F20110114_AADLIA cho_j_Page_093.txt
5b93da8a5f3cc8dc3b8df822cb9f4161
03dde67d89c57f6b161a83e8b2b3f731780e6948
1920 F20110114_AADLHL cho_j_Page_045.txt
7beb70f5ee6487230dc8a0c212624d40
8efc89edc76d3163da0668bbc84b6bd25de858f0
1595 F20110114_AADLGX cho_j_Page_011.txt
62912fd0e4fb08389cdc9695856f2fb2
6a90a6c1217dd073bebb8e9cbff02cebe0808c26
81146 F20110114_AADKDV cho_j_Page_021.jp2
0757812db0133622676d59a56a0d7309
8e367e67c539a7205e2a12998a1eb808beffd34e
19304 F20110114_AADKEJ cho_j_Page_068thm.jpg
afe054d323a6276b310446005a49b6ee
3bb24835daa8be5f2468810cc987fd4c4a9a4016
F20110114_AADLIB cho_j_Page_095.txt
36a58ec159cfa1018812b030aabfcab9
ac3ab732ecfda2d62150c7f5fb7179ba32a1c130
1891 F20110114_AADLHM cho_j_Page_047.txt
889ac659a017bf56da86fff14403fd2b
05a6e0e0b0fd6b75c6e1ae90772799bfe45edbb3
2011 F20110114_AADLGY cho_j_Page_015.txt
613826e3423b8387ddf3216b257f9240
e25e62c407bc6fb795f2816bbee1e49c372431d0
1434 F20110114_AADKDW cho_j_Page_097.txt
92fe96016a5a085724b076630b9c46c1
33b96ed3c37db8822ad896248c0a07ddede1e29f
F20110114_AADKEK cho_j_Page_010.tif
3005e54dbf45f064d1beddce63d8a452
00bab47e86d98ce255915bdee159b5b87ed109e0
1359 F20110114_AADLIC cho_j_Page_096.txt
eaa4f831ec055cdbc3ad92ef0b91a2a4
f078ff82ab858633f75569c36082989f2f9e249b
1576 F20110114_AADLHN cho_j_Page_052.txt
8c0e92cff18711b317cecc647b45785a
7211dacf96b4deed8186d0e45e7638c3de6dd068
2032 F20110114_AADLGZ cho_j_Page_016.txt
46ba0f12e87cc8b1fc429ca3110c0414
bc8df0c6db08e25c40465fb9f269704dc036144c
25320 F20110114_AADKDX cho_j_Page_126thm.jpg
27365ee8bc07bb84a1bd215db8eabdb2
03046dfcd0f917e0d892fae107507b627ef4fb97
F20110114_AADKFA cho_j_Page_009.tif
cb2e309fb494b17410450cb01475bbde
1f67562110fbca94104c4a4c5b269ea2972e39ea
182737 F20110114_AADKEL cho_j_Page_071.jpg
efc1b70ec2a91f0d5b28c2720d52b148
55ef111a03858275993515057560acf7e47acdf6
1942 F20110114_AADLID cho_j_Page_103.txt
205128a101bcbb2c6a19818f9a0d58f7
be36fe99163a850c3f9963736e9330cae9cd5a22
2027 F20110114_AADLHO cho_j_Page_055.txt
fef9eeb516e5acbde4aea0374f778819
b446005fffad2002db82dc48ddd5b71c4e610084
632644 F20110114_AADKDY cho_j_Page_039.jp2
08af3c4c48828f82e1c5270bfca935fb
d44edfee0c52dcb83f3c2309b74b8c34f1179d4b
10484 F20110114_AADKFB cho_j_Page_120thm.jpg
815cf35b626e2ab568d9cd4087aba945
a1827972b0da6c05ced1c6393b48e655ad53a0bc
202725 F20110114_AADKEM cho_j_Page_081.jpg
a7f89230d34e6565521b3ce0912be5da
e76aa34a31fde37b02e735f3f407e76a6398618a
1915 F20110114_AADLIE cho_j_Page_108.txt
9f5c4ac2f674cf2ddaeb0d5802f9355d
aea7d104b28762dcfa772d79a097a94ab66f9007
1292 F20110114_AADLHP cho_j_Page_058.txt
8f50335356bf673d10b0db17fc61b5dc
366e8c8e2b6243f18c060702cbd54a81d63c2a53
240144 F20110114_AADKDZ cho_j_Page_036.jpg
b90e9285d1df1fcddb3c1baa30509503
3948b3e92bf1b13f521fb1555b0ac2ced7447baa
53411 F20110114_AADKFC cho_j_Page_001.jpg
18ae3fb86c1d3efd7d28c11e6c7e5891
919f6143a1c890c044c361933646025bbeebc40b
43403 F20110114_AADKEN cho_j_Page_084thm.jpg
153357ba7ed69259a9daf9a87755d5b3
af320965b02d716098d725c68f72173bf0fba90f
480 F20110114_AADLIF cho_j_Page_110.txt
6b34a2c6f48d56d359a75d2a33305285
51ed1428e241e71b9c365e5019fdbd7b1b43f71d
1768 F20110114_AADLHQ cho_j_Page_060.txt
7263b105c9a612855fff9100445b922b
62d2b30a7d695fb09114d28aba9419840a1e5aac
57604 F20110114_AADKFD cho_j_Page_124.pro
35fa4922be38434f20efc5aac7e60d2e
9edbe8028b9262aeee3b864125a72cc42071f69a
1006357 F20110114_AADKEO cho_j_Page_071.jp2
9afb2264c6d88527d3898a94579e53e9
b72de954eda3a42e293cd04f2bebaa6158f0e083
494 F20110114_AADLIG cho_j_Page_112.txt
ead92f3e8074699c1d4424ac0643d890
7aab5475c3986cae7bbc3c77744726fb015922e5
F20110114_AADLHR cho_j_Page_062.txt
e7302382331b7259fab34322f0fde9b8
39455685a8594ae31c088112a376527fb3f88bd7
F20110114_AADKFE cho_j_Page_118.tif
6d0538b62dcb67b34ea57579057471b4
9d64229900c315bc559c3f8bfe4eaad478730126
1416 F20110114_AADKEP cho_j_Page_022.txt
337c5792410064ef3069ab05f8f5b11d
8e8f5732038870383086fda2e879ff12946bb92a
1809 F20110114_AADLIH cho_j_Page_113.txt
c008ac2ecc68449f8f63ae40ed4bb2e3
7f27b7fa3d7e98db23b1de13b5597ad32e870b85
F20110114_AADLHS cho_j_Page_063.txt
8ad81ede90ae288895be2424a4325ba8
6ab6aeeb868c6b922963d961cf8992d50842db65
194982 F20110114_AADKFF cho_j_Page_104.jpg
6ab4a43eef06f3400253cd61d4e741cf
9d771f6418436773fbffcf03695cd6c2b2cf0889
F20110114_AADKEQ cho_j_Page_060.tif
10d20c66a3abd737cc04ad7a68350b84
6984414e6644db92365ef0782740121bd72be3a3
2152 F20110114_AADLII cho_j_Page_115.txt
1f2461d5c0fc33afd45dc3bc67033996
8d38217b5f2b502ced69c0356b44764cb57175c0
1724 F20110114_AADLHT cho_j_Page_064.txt
49229fb9674ba63a4b4a7e3d87716697
9195261d122c10b045da519d28da6f69b6b011a3
1051914 F20110114_AADKFG cho_j_Page_005.jp2
d905b92c5432dd631179b9e796d11242
ab57246ad43138050620c6ee97d5a8577f232d2b
F20110114_AADKER cho_j_Page_102.tif
561ee93afcaf1d5ef2760f7197ad15d7
b6224c87ef724de99938a03432af6c54bc10d5d4
F20110114_AADLHU cho_j_Page_065.txt
959d14c4266026a646c60924f09529d9
32357a1e70304c73570b1e95795f6a139ece8590
37521 F20110114_AADKES cho_j_Page_110thm.jpg
8cbbb9334f63432390edf59c6e82b9e8
8a9dffec2d15336c1e2149026bb1425db451575c
419 F20110114_AADLIJ cho_j_Page_117.txt
49897d8adfe80b56dfbdc89adeee2d6f
238441ca2cfda913e257745757e4c667fe094c51
1737 F20110114_AADLHV cho_j_Page_071.txt
7f4c38466c2485df87b85ec5688b6fb6
47a29e8180667df869ef3b5124aa550fb7e79ce6
22957 F20110114_AADKFH cho_j_Page_033thm.jpg
f8d72fdd4233ccdc07afce327d2dfd45
75cbe5e8d9e32b52eb952e75826e70dc5dba3004
F20110114_AADKET cho_j_Page_061.txt
16c8a6ca9bc64032b298d61d8a866eb1
726686943d59b3bf6ae980918361bb4fe2175bcf
1454 F20110114_AADLIK cho_j_Page_119.txt
7a4c41a70dc1d4870e49f8e1dcaeb0e8
aafed38a4e24589df68881f63650ae7158e5cd2d
1491 F20110114_AADLHW cho_j_Page_073.txt
53db7cdddaedc890b632f885a6b0809f
c91a80a4255689350ed1f0022062ba17d6841565
79892 F20110114_AADKFI cho_j_Page_040.jp2
4e47d0eb740746914557ac5fd94f5acc
18cf0683dc14c78022fd08012495ab7eba2daa35
F20110114_AADKEU cho_j_Page_022.tif
641d28b22945beb570fc89428c218e49
973047c656ead51dced606fa3c7dca6b5ef86b11
23953 F20110114_AADLJA cho_j_Page_017thm.jpg
18b5cbbed589103c9e622eea89696f85
c3f2388b493c7e098dac62076b58d90748360fa1
1440 F20110114_AADLIL cho_j_Page_120.txt
6ba76c4c15e15000bfc47205d9d7a23c
8678950399432972d81a5e0af7eff8573e4e3c84
F20110114_AADLHX cho_j_Page_085.txt
bc6cd11a4b69f6a8a7a06736f840391b
aeeb064b4fd7ac3072eefbb25eaa7c11776558f3
208709 F20110114_AADKFJ cho_j_Page_106.jpg
122b79af2837b5a3ab57b8d2d6ce8dc0
a7406d4b2c6707c4571d6387ca54e88293d46d01
27402 F20110114_AADKEV cho_j_Page_081.pro
515c6c3d86a6a381434eed7ff53608c7
ba3376a2e408bb88d6b9f85512fc7d7049558415
88122 F20110114_AADLJB cho_j_Page_097.QC.jpg
4fa0378c66541ab9c9edecbd527779a7
db78b2086f627a660cfb69bbe9c92ffd0accc57c
2317 F20110114_AADLIM cho_j_Page_128.txt
b290a27e8f5a0c18f483dad8718c674e
389323d39db5ed62332bebbdf49a8c5825dd6f37
2182 F20110114_AADLHY cho_j_Page_087.txt
d245a095ce462e6be734732de72af275
76398f791c475c03583133387a4a87ef1c212240
71211 F20110114_AADKFK cho_j_Page_060.QC.jpg
1ad9670664a112a315bf3d25a85624b4
d56de39b0eb82242b92d38777c2a66e605d8829c
45241 F20110114_AADKEW cho_j_Page_104thm.jpg
03eea29a998ef1ff22595a33da437136
ea6a923e1f4195b476e2388e2813a478ddda6be9
92451 F20110114_AADLJC cho_j_Page_106.QC.jpg
68b0e571b32f40807b916c40490b0699
0f4c50525807343bf6d926eae4fe1fa8b52f901a
88034 F20110114_AADLIN cho_j_Page_127.QC.jpg
56c75e7ec220d167d488c2ba478ae1e3
0308a5b60e45607385b0d9e8a8508aee55018b49
1916 F20110114_AADLHZ cho_j_Page_092.txt
19d0077704bca97bc3eaac460b903444
5a563e867090bbc3c0c6dfea609d64ee69c7bbc3
48095 F20110114_AADKGA cho_j_Page_045.pro
ce1ec1307038d9321407d6f4eaddef6a
8d23454ed3202451a956d9a33f8f72e98e919229
24892 F20110114_AADKFL cho_j_Page_125thm.jpg
dd165970ffab980535a2f0c214fd0eca
1cf3b1c097829b941cc458002936591ca8f91844
1762 F20110114_AADKEX cho_j_Page_033.txt
860adc8b1944affd4306059fcedcb477
50b8ce895f79cac9e1ad47ff8ca523cd17bc0e86
26476 F20110114_AADLJD cho_j_Page_016thm.jpg
8f33560f44bb0dee33a7da68a19978e5
d23315ca6079b24e3e36fc2620e817e0f0ca3c73
24786 F20110114_AADLIO cho_j_Page_128thm.jpg
4b0de1a60b59e989423dac799e8f7311
654aad3377bfc22846f78f0346987086fbb52f7f
96029 F20110114_AADKGB cho_j_Page_035.jp2
6766d9e5e0bde53e9844440bbc240e05
e542d8d013f6e40487e9d2a007ea80bcf14f3bdf
1712 F20110114_AADKFM cho_j_Page_076.txt
f5513454c9f79c02b49033533cfabb57
c7a78b82929862cc693077e2c138e3241d3b271b
98061 F20110114_AADKEY cho_j_Page_047.QC.jpg
c79ec1aa6a5833e549fe328e2a698897
4ff78e650f9ec79a32af5f99c7105416787a3f6a
44348 F20110114_AADLJE cho_j_Page_079thm.jpg
67f7102c8143d52d24da9b5405288a67
e17fcfac616984b17d1bc3c9c266d795ee3a3811
78006 F20110114_AADLIP cho_j_Page_048.QC.jpg
b333f4dddb0a0d5750be103997da242c
48dc902cac3c5294ab807cd7eebc3180c8973104
192957 F20110114_AADKGC cho_j_Page_075.jpg
7a16da4d290ea1f5159a24b4162da32d
dc1c89f42074bc20b532f0b3aa53c1a3e9950670
F20110114_AADKFN cho_j_Page_029.tif
d2fded8fb3042668987a6e1e7208e928
7e515364e73862a6ba82746753a81bd06a54a97b
F20110114_AADKEZ cho_j_Page_050.tif
56263ca27feaf17cc500a668ceb5013d
c02ee2ed11ae33ea60459310a69f94d0bc9b8274
60080 F20110114_AADLJF cho_j_Page_069.QC.jpg
32d0aa9063882bc1d55aad013a36019d
fb8aceca7f7b57ee7af6a6c86a60aed8e6230951
60547 F20110114_AADLIQ cho_j_Page_067.QC.jpg
6b13fabab9876a0bcc308c6466905b90
89c98e6e5ebf5934ac2f740c571ff67948cac76a
31079 F20110114_AADKGD cho_j_Page_100.pro
b71664c03236e470d62551b4bd1fc477
c3c0317117ad9cf91a73e2b96b811317e3664ebf
F20110114_AADKFO cho_j_Page_130.tif
264292b32053ab6f0469c7822ed0e7fa
12db39a05e323a414f1cd5cc2cd7fd49e8c99f8f
23965 F20110114_AADLJG cho_j_Page_024thm.jpg
7f6349ef5dfe013e0f4364205b5921ae
1632bcab3b22854da46bd199d2c60dd81e725a1b
42094 F20110114_AADLIR cho_j_Page_112thm.jpg
bf42f0516651ab75748c9c192f63523f
1cf40de8c5c219f6af550cd090c4878a027a979d
62063 F20110114_AADKGE cho_j_Page_058.QC.jpg
f8b209e5fce72847fce4a62f504e0d6c
6ce4f45920427058565bb885a4414586bea51e17
1868 F20110114_AADKFP cho_j_Page_030.txt
be28aa6e2d3cb43264409a97bd34751c
450f9f8d357bfe63be58637df9b5d12302676de9
4041 F20110114_AADLJH cho_j_Page_003thm.jpg
20597d27d497e9dc16286df66b641eac
abeff9ce93003653ca9d59a64f252bcc8deb9151
77785 F20110114_AADLIS cho_j_Page_114.QC.jpg
0738da644e1b317585c22b1ff048153b
2892b15d22e0a6490e65348d95f593b7405f2f41
F20110114_AADKGF cho_j_Page_017.tif
59af222d8714376c963ef2fc545148fe
07bde65108842d5f3d77124ed16875bbf246228b
59211 F20110114_AADKFQ cho_j_Page_098.QC.jpg
1639299aca4f8569f911ced8d59d2454
b139ae03c4debdf05807c8dd1db81facc9889901
45458 F20110114_AADLJI cho_j_Page_091thm.jpg
1b14e272c5c05cc3c4cd4a43d6d5c0ad
9f2fa37f9a7bdd7b1aa50f83db6d48ac3de2f27c
80383 F20110114_AADLIT cho_j_Page_079.QC.jpg
4bd2ef373199899dc013057b0f58cd76
bc043e73c0808fac1206d74ab6f694120a35c2b7
25376 F20110114_AADKGG cho_j_Page_037thm.jpg
21791e45f47fbab763a985fb2d320e6f
e879c730a58c84d23fd7ab9b0454913e4b66d68e
114493 F20110114_AADKFR cho_j_Page_044.jp2
b03e5ca3a7195a6187dc1eb81a457fdf
fb9d1a827a2cbb43769d3f231cedc6e3df5a4a35
83160 F20110114_AADLJJ cho_j_Page_016.QC.jpg
e92e37651a7e740f2336f1b7bfafffcf
7cdc7f616f396f1e8dd2ce73e32dbf076f63e488
124055 F20110114_AADLIU cho_j_Page_010.QC.jpg
791e002b798ec7d2fa18e478f1b6958b
c91c72064c37d9d9db7d03b8e3880a1c1dc96973
1051975 F20110114_AADKGH cho_j_Page_102.jp2
f4fb1ace7ffafbd3253eceef9125dd98
587306c347aae547adb491f0c5f5dbca07b43f68
116718 F20110114_AADKFS cho_j_Page_110.jpg
a7b858c95ce8c0e8827c4c80eba525e0
1cbb785a7e414b39b165caef5b478309632beca4
63324 F20110114_AADLIV cho_j_Page_022.QC.jpg
cc048dfcbf3c704e7e8884e7de96828d
efbd444bf78821a9e21c574d7b7545b13cbd4c80
30724 F20110114_AADKFT cho_j_Page_104.pro
c5c10784cc63f9cf84173f4100f8d71f
4282cc8aef0dd2272423b8233bc53ae64f2cf682
50290 F20110114_AADLJK cho_j_Page_047thm.jpg
9faf1eee02c3601f36b88fe4740b9080
d0ffa38ce64c97ebe32ca1280dfa2ddebed50c0f
27284 F20110114_AADLIW cho_j_Page_043thm.jpg
f97e95b7ff88838514ba76ab6c3e4ea6
6d8b493f170460568c4649da9426d9929863687f
F20110114_AADKFU cho_j_Page_079.tif
54e07d86ecda1e0bfd9b0f513e508c03
5cd55da5d01f008716554d9a6d18d5483eaccafb
221556 F20110114_AADKGI cho_j_Page_007.jpg
00170657b449b9b022d30acdfb9d515f
9cb33821e3cd79a21521f45259a7f1c024fc0bdc
72769 F20110114_AADLKA cho_j_Page_023.QC.jpg
09294a0460934f4a0af810ba4cb4a264
56cdcd3737321092aadf7631a76509f06ba04d4b
51363 F20110114_AADLJL cho_j_Page_036thm.jpg
8367ddb6d996094e48b63efbdda2e010
37cf4df7942c7ba3639765e72b28847d64a6c57e
87144 F20110114_AADLIX cho_j_Page_099.QC.jpg
8f637b7ed4e7302ec85e2d282af8bc20
a10bdac681ddfc8f22c6c5cda01b49134ea22a3f
F20110114_AADKFV cho_j_Page_061.tif
111664cda3c2920beb1b374ca0fa7317
2f0588104a2fb47804ed8062aa0e5c5f98b6ca83
17029 F20110114_AADKGJ cho_j_Page_019.pro
4b7c9f249e1593979d98582997ed5469
c8b11cc21b67691f940fd210858c579d5fa40f34
27560 F20110114_AADLKB cho_j_Page_025thm.jpg
f3947d8ea5713f651292eb3547cf0e88
63c913fb91867d36b632fbf04d388be715d820e1
25052 F20110114_AADLJM cho_j_Page_124thm.jpg
ef53cf5011419af5e25c0999ed9465f4
59631302f1b3a6df0f0b376ca918dc3a40046356
67705 F20110114_AADLIY cho_j_Page_111.QC.jpg
754a2afe32e9e3cb5dcb3e01f9308d55
dfae2899a19c1d2fd45fc52903db4b9bc3e64f08
F20110114_AADKFW cho_j_Page_089.tif
51823ed573ca9dd4246e9253ec74b4fd
fb5db0c1910eba563c9e0166ac465c9109eb01e8
40218 F20110114_AADKGK cho_j_Page_093thm.jpg
3c6fc9ef1b293c84245b029a045747d9
df0c2b6072b2211df3f54eef4a69def847345cd4
29525 F20110114_AADLKC cho_j_Page_034thm.jpg
91fb13aeecaa4ce4884d64f4dcfae4d9
0e2e019d732021731badea3c1f0b0af6d01352f4
96494 F20110114_AADLJN cho_j_Page_087.QC.jpg
d0b4c15de9fe0ee948dbfbee8d4dfd30
e688b1d5f5fc85e4d103cf5cfbbf626a966cafd3
62647 F20110114_AADLIZ cho_j_Page_093.QC.jpg
5ca9a7b624a7391ab0912efc526c6284
5bfd436137fa5343320b8d386e1a0013a007d991
134818 F20110114_AADKFX cho_j_Page_093.jpg
f9ea45727aa3d460c1a8388077183930
a4b449ba515aebc0bd078a3b6d0427a10cb5f16e
39203 F20110114_AADKHA cho_j_Page_069.pro
a005204436af03d9779375f85fd25837
ec5ed6e171e9c81f86d56ded27a43a66e28add91
710 F20110114_AADKGL cho_j_Page_012.txt
4a359b6edb441f04f19601d663cd97fc
c4d2fc3945f477a458689498f7d2f97b17ccd5f0
67321 F20110114_AADLKD cho_j_Page_035.QC.jpg
64f590fbdc401978cdbcc17fbbea22fb
2eaed650b80ed2f6f4577d4ca47a69307575a708
45005 F20110114_AADLJO cho_j_Page_005thm.jpg
968b8eb40ee6328e34556e37592602bb
e359c085d670ed6e9c3b02a11e96cd8ba2daec14
51234 F20110114_AADKFY cho_j_Page_015.pro
7247c6d595bdaeaa9058ab95f72198a1
e5b452cbe39f42a31a3a7bb44ba9ff5d3d9de6ad
F20110114_AADKHB cho_j_Page_054.tif
28982b5560941116dd86c7189f814c9a
6ce6942fe251a7fec8779e308bdb669fa9b1ce55
634791 F20110114_AADKGM cho_j_Page_058.jp2
88c22f5e80e08e653362311246cafe95
d191509d60e68d84c0a84117ab4cf6f207818e9a
19402 F20110114_AADLKE cho_j_Page_040thm.jpg
84ea70c82f86843dbb4c638c3c8344e9
e07d6b42640329d2c7ea38abd62435699ed22d3c
84253 F20110114_AADLJP cho_j_Page_044.QC.jpg
c86bac014e9de2b9eca2a72a6999135a
ae1b85ca316de66c8c08869726547d41f60e7c5b
200715 F20110114_AADKFZ cho_j_Page_092.jpg
71a1f1e08a426796c42344420eb8e37b
e81d7b5b094bd9ae8851b34f1e5920b644c18bc8
F20110114_AADKHC cho_j_Page_053.tif
c049b6bc18f2a3ebcaaff4b03469113d
f1b4fef8f6d81b24d97988bd83cd88776607c79d
86684 F20110114_AADKGN cho_j_Page_090.QC.jpg
b0915c4a6bd8e67e2d4efb5ce6591eba
e1b0b820b2d83984212499c971a8b2c2193b7d7b
23388 F20110114_AADLKF cho_j_Page_041thm.jpg
946f4d463cd3957116ed10f7cb1d88ab
0ff74c11ae52d3a9d2c95e7e89f8a5d284c35277
72015 F20110114_AADLJQ cho_j_Page_020.QC.jpg
5ddc728a76830b33a4daccba6c49f8c0
c22fcb1fa393f2a82d86a8f514cb0cc29411450a
2467 F20110114_AADKHD cho_j_Page_123.txt
efd5fe7cb5d2b8b5234d5a179b72f7b9
2c1e6db44005a73d682c256df40ad891132b4a64
26960 F20110114_AADKGO cho_j_Page_120.QC.jpg
de071782d30e92467b83e401b45f7bb4
68367d83c71400a768533127533761f6fe457115
26058 F20110114_AADLKG cho_j_Page_045thm.jpg
0d3aae95d63f6722174289d943cde385
b1abb582cd165bd8dbe38537641c4bb58492cffa
72891 F20110114_AADLJR cho_j_Page_041.QC.jpg
5bd85279a66db2f8b0c4b0f10c264d36
fa85fa198bcd5920074b2ecfb25fbb25dcf6ffcd
F20110114_AADKHE cho_j_Page_052.tif
cf57fb4179bd2f2b6e4450713c52df18
d995a92a0baaa21efb464073697a35b91de1e99e
57779 F20110114_AADKGP cho_j_Page_126.pro
a8bcd7181e8500d4da104cee7dcaa073
80d822ea1571eb28b6304851704bda11dffd00d3
29880 F20110114_AADLKH cho_j_Page_050thm.jpg
7b812c10e18e88996d1d41fee00b34b8
3c1fd6593ea7a85bd8a3c88631a08383202bf20b
192301 F20110114_AADLJS UFE0008180_00001.xml
5ea4ab89d4a2f15105e23e892d622b25
469e2d3341d6178716b6449847285a9b4ef5ab46
F20110114_AADKHF cho_j_Page_048.tif
89cc5c771653c5826927628b0f48b28d
13628e3fbfff74c300e3d3758cbb1ea907a9a279
285901 F20110114_AADKGQ cho_j_Page_122.jpg
bebc2540946824c2bb95be0997f045bb
24ef642fc0f7f992dc2b3b14e681ea7a8136b0ab
65217 F20110114_AADLKI cho_j_Page_051.QC.jpg
ff9d4a4214dd7af1f202df84e94e3864
c4d14be22bf188e2dca8f5c9148bbb9862c3089c
5845 F20110114_AADLJT cho_j_Page_002.QC.jpg
793b1b13ccf22be0f897d3315bc62ddf
2c61d81e5e91690280c39e583280ad917e316367
51408 F20110114_AADKHG cho_j_Page_055.pro
042bcb40162696422e9ac17d223d10dd
a2ce9c533d735dad870fe8cf40d8f2f73408fea1
86062 F20110114_AADKGR cho_j_Page_076.QC.jpg
fe57c07e92e815c04ac94fdfcad910f1
5f8197af44d35cef29c4a33f998ce19468725eca
66254 F20110114_AADLKJ cho_j_Page_052.QC.jpg
7468147fa2181db183f02d9e6fa3dcfc
780bd13f2c9f4fe8ee94e714f9ed9d3176449a63
129197 F20110114_AADLJU cho_j_Page_009.QC.jpg
cdef62fd3cfdc8645e3251132a033680
ce59d6721dc7188cea7e320ff6a5d66b7f635d49
816552 F20110114_AADKHH cho_j_Page_079.jp2
d5ed438a70e43c23732f811530951430
42af691f2583f7a9346dfa904bf2765bd528800a
94524 F20110114_AADKGS cho_j_Page_113.jp2
d263fe9c8cfedd78e9389a86a16a8937
7ed689bb54271868508a4ae523419a728070db06
50401 F20110114_AADLKK cho_j_Page_053.QC.jpg
85f4892b6a766b7ac7b70e3898fbf0ec
585de27a528e80cdb7a87f4a4aca8554ddc2efeb
25794 F20110114_AADLJV cho_j_Page_014thm.jpg
32a2d1c8d62bb279a0a295fb5d0d3a7a
f9c30ad6339abebc224ab39bfd7c43e0c4455dd0
90723 F20110114_AADKHI cho_j_Page_081.QC.jpg
513ffa1cbe0e0f2d047abd69bdc921d9
3488f2dcd27adec740aa4d038a476c8ad05d0fee
29555 F20110114_AADKGT cho_j_Page_089.pro
33b84a6f4fcb9d28094582c8dde79782
f592f94bd02f9a4c04a71d8ae1c3872662240484
26274 F20110114_AADLJW cho_j_Page_015thm.jpg
13bc4122afe1756a8026991ad1277c14
5274c4f0bfacf05027850add71f2672806146171
983867 F20110114_AADKGU cho_j_Page_101.jp2
bef8a3ac94d5b5a18cccaae74573acac
76a0271f552053f6b55268a255ef7cb44139e204
20044 F20110114_AADLLA cho_j_Page_098thm.jpg
815f663943c5844a83c9d1da8aec2e07
eecec4f90d186d8918bb964790bef0791d7c69db
49749 F20110114_AADLKL cho_j_Page_056.QC.jpg
5784afd635f92f8e311e098c99ad8938
61238a7d5564a5d755c4c7f4fd62d4709b36e052
25358 F20110114_AADLJX cho_j_Page_018thm.jpg
a8f6deda0ec9c5daf51cf6db6589e5bc
49de6811c9f73bd1b8e743c374097709c7eff1df
1672 F20110114_AADKHJ cho_j_Page_118.txt
c6b028b78d20e1ed12fff86d7585a008
d8e54224293bf8b22dfbb1191da47703f9a776ac
162596 F20110114_AADKGV cho_j_Page_069.jpg
7588701a8fd6828020f937371048b94b
c8c8e2aa4e4e9b716260d1cc46890d54d7762a0a
45015 F20110114_AADLLB cho_j_Page_100thm.jpg
81d37813bbe06ea0baf0e8339a35a599
865b0c03b2e4753ee2bde8ed73d6080022ef93d5
55472 F20110114_AADLKM cho_j_Page_057.QC.jpg
8c1e4132289e1d83019d466dd140d8ed
ca12d35524285b9c7cb8b14406d605889b295643
30375 F20110114_AADLJY cho_j_Page_019.QC.jpg
f6e7c58afc3c983e513e841b8a9c44c7
fc6c9a3c79100ccd1772d27bfa8ffe390d72d3ce
17705 F20110114_AADKHK cho_j_Page_012.pro
f1f90c646e95b864a7f3e03135284908
7b9952b00be170206c4154d403571faba03c494f
1688 F20110114_AADKGW cho_j_Page_057.txt
b539ddce63103fc6a51c01791db08772
ce082bc9e60d6d16deaae3141bf88cdaef8995a5
86610 F20110114_AADLLC cho_j_Page_101.QC.jpg
89496f8c6c6d15668b8c5ec7c2e0f7ec
1ece1dabdcc34657ff23450bebd7dadad5201a8d
46038 F20110114_AADLKN cho_j_Page_059.QC.jpg
a0add5efdf1ee55086124cec3f5866a3
8b61c35007bb7d2a3c2a540a20b106d2d50b4f04
22710 F20110114_AADLJZ cho_j_Page_020thm.jpg
9d52dc63ca539683e3c8484124ce89c1
c5e93c860cd717149d350cedf8a527b2537c10b6
183733 F20110114_AADKIA cho_j_Page_063.jpg
4c57e72fc8f1a1da762fa5adebf3edb1
e2551506726fc202ceaa480acc198267c3dd5922
55019 F20110114_AADKHL cho_j_Page_008.pro
66f2be797974d24ee8b4b462438fda63
a6393d5c8f7e4bd66b2269e899502a9b4573ee87
F20110114_AADKGX cho_j_Page_026.tif
5e80a44107a727dacfd96be2b5f0810f
b025ba426aa162c090cf9766364dacb7f30a6d68
23026 F20110114_AADLLD cho_j_Page_113thm.jpg
20f61f905e69d690611afc8543600d33
869f85cf6783f9996e1e4ef66dd51279bbc97f98
20774 F20110114_AADLKO cho_j_Page_062thm.jpg
3593d22c738f65fb020d1b4ff5e955f7
fcda7da9f777df5ac0aef69df5893c0cfc2a7d05
96950 F20110114_AADKIB cho_j_Page_080.QC.jpg
74402c138f230cbb68deb9928b698c7c
5c520e72650a323bd60c6be89d198201619b3b06
1446 F20110114_AADKHM cho_j_Page_059.txt
2a3d7ec9636ec7843af16fef3c69b605
732989ad7920ffab97c53866ad9ac417fd0c4ddd
82235 F20110114_AADKGY cho_j_Page_089.QC.jpg
06b5a53d79be65583337577b21f45ecc
21a298821b93f1ee37f421522bf482ed34ae362c
70307 F20110114_AADLLE cho_j_Page_113.QC.jpg
8536eeea267899c4f17ddd9c71ba535b
d187136e87b318ed1a8213b0220bb75143bd82a1
83883 F20110114_AADLKP cho_j_Page_071.QC.jpg
a26e7d0928b2d8928e9e724f6061d227
d15e1c74ba8e0b748a86276a4015e631111af421
161074 F20110114_AADKIC cho_j_Page_051.jpg
f32ccc1ca17ccd6af7e208a9b059700e
17fed8963773e092c0abfec705317cfe112ff0c9
140666 F20110114_AADKHN cho_j_Page_116.jp2
12aa4d83575c80bd60d3b34f8da51b96
96915ceda2aa0058a767467d59ed0ac48d0f4cff
F20110114_AADKGZ cho_j_Page_067.tif
1ba710d37e5ed7a4c8253be098e1a651
f842a598f85e02758afcf86b8bbf54a3e31dcc4a
24668 F20110114_AADLLF cho_j_Page_114thm.jpg
c344251fab1b77be03f07b3056a59411
67b655155c965a56aacecca06cb208d76de2886f
94207 F20110114_AADLKQ cho_j_Page_073.QC.jpg
2a909b45068b2eb3eda4203faeaee3af
e11105670e63cff16c647d07804186852d15e713
1820 F20110114_AADKID cho_j_Page_038.txt
a2a8041dec8183e82cf2d73e27119953
635c1b3c18d7c5591a44f01fb37a43d62b914986
36808 F20110114_AADKHO cho_j_Page_119.QC.jpg
a881e772c1e5992575392daef71d20e0
35088f6d33dd76f4870a7edea9edc753a2f469a4
25671 F20110114_AADLLG cho_j_Page_115thm.jpg
5523a5c2b9a06f6e7b84f34200cd990d
2c8a7d52fb354ee0247eee7c0b01897d4c99ce0f
66694 F20110114_AADLKR cho_j_Page_074.QC.jpg
ea1171adbc707e741f7e76bf9b8d7d62
de55373d69a4205f79573c03d940c02c9a709efb
745 F20110114_AADKIE cho_j_Page_130.txt
a95d54c4c8787ebdda6746159cfed282
c5126c6bada5b3a72a8d7fdf3fa650d8180c24c5
F20110114_AADKHP cho_j_Page_037.tif
3bb81eb3a4cbe378d95bd36ddb99f2b6
68180a078276ec29e41b2ea78a59726117c0429a
26557 F20110114_AADLLH cho_j_Page_116thm.jpg
b157194dada986d114abc21bfe51c2fd
b103ec579a0970ab4ba32ced5721efb7411bbd9d
93931 F20110114_AADLKS cho_j_Page_077.QC.jpg
60c0fa957e5021181ec7af0f5ebfd351
3bf87c55c74310a438cb9d47f3e27df49d2cc202
193167 F20110114_AADKIF cho_j_Page_048.jpg
4fc53b8b6263780408436ecc593a6da7
7e1adc3c9ef28beedfce1a483e600d4a0a3f4ea4
20736 F20110114_AADKHQ cho_j_Page_121thm.jpg
2f53232506cadacfe564fbc896636db4
0ca445eb5713c62331ae7d5dc552b4aad753e722
18838 F20110114_AADLLI cho_j_Page_117.QC.jpg
ff4404d14537637dc0c97fe682712546
d38d31ac8c6a08fa9e29581f7ceb060933a72bd6
44862 F20110114_AADLKT cho_j_Page_083thm.jpg
9e94a8d7e79bb850c5b698d6648d1c9b
fd3bf17bc853115a81abca30bb9236fdea1d0668
83602 F20110114_AADKIG cho_j_Page_107.QC.jpg
e94f7e07f7e9b30856a42de7018b4e11
e153a32472b4fa22229cd4bb0bda4e720ee74018
26434 F20110114_AADKHR cho_j_Page_129.pro
3578cdad9c80c4e9a7cebc8dfbbfb3b9
7d457acdb31c3ab1a88b3779774de318b300f2c8
13550 F20110114_AADLLJ cho_j_Page_119thm.jpg
0ba943c8b8cbb27dbfe929b8d55b29da
99a8704bbf89f8e89c71696d1fd9567e0b0dafcf
74044 F20110114_AADLKU cho_j_Page_084.QC.jpg
496ba70370ce80041d285bd64fbb5049
a2260c59dbec3191e301520e539ef07173d62eb2
154589 F20110114_AADKIH cho_j_Page_057.jpg
7bc02c3c5c56c83c7d70f52ba9458f71
e84bd71fd59279653c9a868bcd7cddc3b18a2b2d
103614 F20110114_AADKHS cho_j_Page_042.jp2
a5296f5a683edfce1636361006073501
e30dd8f55f26c21744263ab9e43869594eb10293
26742 F20110114_AADLLK cho_j_Page_122thm.jpg
1606d6f7e624a0376b13ac8bed7ba673
59db7bd446018fdea2de00d98fd993790c6771a7
49321 F20110114_AADLKV cho_j_Page_086thm.jpg
67cbed798214126f51d7cf72391c29f0
06196142882f4696d978beb5832da080c723f75a
87192 F20110114_AADKII cho_j_Page_029.jp2
602ec2b472ffd6a90b9631da95023a71
4523d143a664f89c0ea1b7bea71ff867b35b7611
2713 F20110114_AADKHT cho_j_Page_122.txt
634bb05611fe446c6949c6ccc89d12e9
44c2bb4751618ed81d0bbbd23a1e1c7a891d26d3
25331 F20110114_AADLLL cho_j_Page_123thm.jpg
cc66dee58ce4373d0c451ae5c1bb0d33
839fcd79d164bf31aa55161662e7e4301fd35b1c
88780 F20110114_AADLKW cho_j_Page_088.QC.jpg
c67372bdefbcc0ee7b726c132abfb9ed
92a5166e9d2550e2409aa8e5b8cc7a52c132ee37
50232 F20110114_AADKIJ cho_j_Page_036.pro
a939b5da76fc92498f4291d1e108ea22
17b68064dcde34e3f1aac05495c8723778cdd0b9
44785 F20110114_AADKHU cho_j_Page_089thm.jpg
b7fcb5f9d0d40d6a892c33f966dfb5d0
c1ebd34e7e26471dcc6d878dd38194c07b88b496
89062 F20110114_AADLKX cho_j_Page_094.QC.jpg
8dcbfbbf6ffd0736320942197ead8f55
759bd0f3be8914460ae856a951e54fd8428a68e8
18593 F20110114_AADKHV cho_j_Page_053thm.jpg
c36ef842118e644be3076563ae85695d
6d111165905a0a645249c074d46f26f8389bb18d
11916 F20110114_AADLLM cho_j_Page_129thm.jpg
9231dffa75837936edee63b4c4e87895
5be56df3c63d5c3dac14d9f1bb8506d9452bc87b
21241 F20110114_AADLKY cho_j_Page_095thm.jpg
6ce93b75b22d2b2d73d069dfcc81b0d6
2f9c35e119cd7d39925631d4f14c73dab122fbb9
176691 F20110114_AADKIK cho_j_Page_034.jpg
af778b5ad2b87e8b97db733fa55c6bec
477660a5b10d957745ffba71eb47f5fa21a8b4ca
2378 F20110114_AADKHW cho_j_Page_126.txt
54147787d2250e8f4c745af458cbcc68
2a867dc9f113de9246134a279db84d6e5ac8da31
42149 F20110114_AADLLN cho_j_Page_129.QC.jpg
2558b2fcc0b91c79bd07db582c599334
76910cc6e3825814bddc75b8e77add4664ad5eaa
91206 F20110114_AADLKZ cho_j_Page_096.QC.jpg
b47923bb2d6715c35937afe449c673aa
bdbc7fb7026e7a3f268bb690963e96a38b60367d
1876963 F20110114_AADKIL cho_j.pdf
b6d3e72a567dd080fbfe3cc0933308f1
f84daf1e418fba78a10b19be9b15c167c580ee1c
92397 F20110114_AADKHX cho_j_Page_118.jpg
ac24a48545c138649244dd40530801d2
96caa55c3d81afcbbb6953f360ca3f64ce49d395
54319 F20110114_AADKJA cho_j_Page_128.pro
d55c8777396793124775badc462c4426
cb60bac36e9c250013f3ab8e7bf3eb706ca00d61
31879 F20110114_AADLLO cho_j_Page_130.QC.jpg
e440c8667785c1fc861c546eff571400
7a9871fd1ff448f65cf73c54ac133ae0ba837a46
71257 F20110114_AADKIM cho_j_Page_056.jp2
92e420a8fcc14749e8a1a60297e7c091
b148af6213dd73633188009d58ec73115208ef7a
35698 F20110114_AADKHY cho_j_Page_098.pro
c2cf2063edec327e33355e9dfd8c78cb
ccb8a3c882ee3c714efacf5eadb364ac100289b6
204728 F20110114_AADKJB cho_j_Page_031.jpg
f9262f515db69a609f732c743ff0d091
4cbd59ede9f51d8fa2e7c4d68f488a5add46d22c
190023 F20110114_AADKIN cho_j_Page_105.jpg
1d71e82296aa12e28237d53d72ff883f
3c87a2858561e55a2c7f583d1118cc35dba81081
1051947 F20110114_AADKHZ cho_j_Page_036.jp2
9e3192a235df78fa1713cfcf459b4087
85a5142b718421e5573cb17007b91c9e382fe3f6
99351 F20110114_AADKJC cho_j_Page_006.QC.jpg
1e10dc41c2a9035934b5aea8982ba5d5
4527fc920f996d0a24da085f2254422e7a166a18
209908 F20110114_AADKIO cho_j_Page_047.jpg
cf18ce18dc483f22b8f48c921f437017
92890d0f1cb1ba60fe55f6b2dce378c3eadd6502
92062 F20110114_AADKJD cho_j_Page_033.jp2
eb3c3dafe48797c9506bd569cea6219e
80f38334d6935cc67969db2562acb469c8727a63
45692 F20110114_AADKIP cho_j_Page_101thm.jpg
089bdf54a3a022fb1ffb41247d407f55
9619771914398ddc6d83d60aedeea86002a67e7b
45179 F20110114_AADKJE cho_j_Page_107thm.jpg
1b8385ebfe7ce67419bd0ddb50ec9e4a
82603ea6ebaf03ea5f53315958cc9cce2ca1cacd
44749 F20110114_AADKIQ cho_j_Page_027thm.jpg
9f1713e99744c9e87a9b55838c3ddade
bc1d4cf608b74e9ce58c863b6c63967549412fb9
F20110114_AADKJF cho_j_Page_076.tif
10987da392388577c5a12e2d3335bd15
9654ebec5134e0deb0cbb1f9f06739fcafa55216
F20110114_AADKIR cho_j_Page_034.tif
29faa863aa79bd0dbd3fa5d131972311
4f3ebf749db1462fc1512c2ded25fb27a6c8d4c2
2839 F20110114_AADKJG cho_j_Page_116.txt
33a5f3ed6dacb79f14b4a818d149a51f
3c0feef833cad603b5c12afb12e40799f13bf71a
24768 F20110114_AADKIS cho_j_Page_092thm.jpg
0231baab1a25930ff43be875a687a788
f5436156a2781b3609a0a6b181d982fd069e2601
40920 F20110114_AADKJH cho_j_Page_082.pro
70dfde4ca8c498772a53bccc025f3038
8206541106366c558f7ff7ce033c741352da893c
957068 F20110114_AADKIT cho_j_Page_096.jp2
1b194f648811267d9bb3c4d0c62663bb
7babd747a1497c93d2d3f4f30e73aac1d5d97eae
45402 F20110114_AADKJI cho_j_Page_041.pro
5c2fa9761f9b8dae20a1c79e86dfd0bb
09cafee8740ffddbd4e8b28e4d6ae2df4ec6182f
14705 F20110114_AADKIU cho_j_Page_002.jpg
953d0b2f1b839c1d995ebfbc65b2e3e7
5eea39dc8421e712b650444552ba0f75b904de02
F20110114_AADKJJ cho_j_Page_083.tif
fad1dfe67884e185d39a00a561547956
92d4c62f6307cba0aff58f70813a257db168cb03
88788 F20110114_AADKIV cho_j_Page_125.QC.jpg
fb898820d3c6313bc606480f3d090ece
36924f111be54980efc04d733f3b0392bf23443b
6924 F20110114_AADKJK cho_j_Page_001thm.jpg
653960387628d75c93056513e4151a91
bceb2a639f56d0c8c13d1c078d312493f0041b3a
95905 F20110114_AADKIW cho_j_Page_121.jp2
ba5f098306140fada8976911377930d2
d9b6b9a65ad5f4281100e98b63c94af2d9bd829d
7971 F20110114_AADKIX cho_j_Page_003.QC.jpg
f504294e47de15750ee3f630240e2010
a3517555fa199762a5628ff0d967ca07f6f5fd9f
148839 F20110114_AADKKA cho_j_Page_053.jpg
9df3955270b7ad4de19c517173215a01
f1be781f27d208bc8f6552d0b59b5b0c81c05c51
173988 F20110114_AADKJL cho_j_Page_064.jpg
a2452f0d65e06eb6d8f315d546b90a91
0116b61d8d7727c78d426de55e99d84c70a8a7ec
F20110114_AADKIY cho_j_Page_102.txt
fa788be6a02a4d4780c199044950cc2f
71587169301f2706c30179a13a21cdf248b22863
77044 F20110114_AADKKB cho_j_Page_045.QC.jpg
3b4ee7aebe881eb75a28cced2a74d008
036e046f7fed37642ceccc2c48a19793da6ebef9
68544 F20110114_AADKJM cho_j_Page_116.pro
a057b596234d0807a9ff2e68f4e1da6e
db94280402bf60a25de1422f2fdcac925baf7e62
33868 F20110114_AADKIZ cho_j_Page_071.pro
49a750edd940825b204c61e3b441e52b
cd4f5da7990197e3b8e878a98a85be6f32a0b5a4
35257 F20110114_AADKKC cho_j_Page_011.pro
e6f421219b966b1ec020d2a1d7f35385
e31fc1f5d476b91db752b988af6aaade662485ca
111853 F20110114_AADKJN cho_j_Page_015.jp2
64b2cc9e278178f77613c59ba74fbe8a
5371d56f30f73a01f613796d5baecc422d3ab600
21921 F20110114_AADKKD cho_j_Page_013thm.jpg
d0e13015915b9f36ebda08764e2e2235
9fef09b10c7244a3bdf6a4d5c2fc62b25db8fd27
1029402 F20110114_AADKJO cho_j_Page_077.jp2
c5a2df7eeb919f8ca553d7190d95e28f
6c1fc5697dfc86be0798b255cb0ff2680d499158
1620 F20110114_AADKKE cho_j_Page_099.txt
5462fdef38496d0d0f0e7a19982e3312
b5736ce634c1585fd7ab8f878c3e5d77a46f5f17
99344 F20110114_AADKJP cho_j_Page_085.QC.jpg
b36dd410d498835a71d825d122c47cec
48efbf4779da2ad9722e6aa21b0baf5bca3e5f96
179372 F20110114_AADKKF cho_j_Page_032.jpg
aff915f93d3ac659b73db7dcab1d0f18
cbb5271762174e2c70d2a8c22f6d02d1e8a66505
F20110114_AADKJQ cho_j_Page_115.tif
5591260417de1f9f56dc7d50c02c2265
cfe5f455076c558c19efa9f997aaf3021b07b66a
1909 F20110114_AADKKG cho_j_Page_075.txt
00a51325cba347201e2c1e609a80d68f
9e51c04c9b4550536895e59634cfcc0c52db7efb
1886 F20110114_AADKJR cho_j_Page_023.txt
26fff3e3ffb0399001a11bb19d7b406a
3378e0eef6149dead7357ad62d2cebed1043b97a
7316 F20110114_AADKKH cho_j_Page_117thm.jpg
2a9ca89eda7ad91490356c9067e45b21
128b758fa903c2e516dcfd98513f6cd18117b78b
46367 F20110114_AADKJS cho_j_Page_088thm.jpg
38b61bd7656d337c477913ba2e35b07f
d830d39ac9eef5e313aa7dfa5a67479c832249e8
77774 F20110114_AADKKI cho_j_Page_042.QC.jpg
2c83d3d94af4358976c5c74062f6982b
1789d39afddb5de1646e0de303746144e4dea244
F20110114_AADKJT cho_j_Page_023.tif
bc7e32fe0ec39acc86f045a72d3b6189
3216daa673c286e69c9e96fee1ac2b68acbd3490
F20110114_AADKKJ cho_j_Page_042.pro
440e127854555046aca96c1f82e434a6
cc4d849fd593c3f46c17a4befede46914e2bde32
F20110114_AADKJU cho_j_Page_040.tif
cf245cd4c92896afc1ff655a52cd086f
dcc1d604f7172efe3bb262ea88774471d7b3630c
188712 F20110114_AADKKK cho_j_Page_103.jpg
51b22261627ec9286acf4017483e99f3
a9b4bab98c98a11d44a140a5395ab55ad0adbae9
10609 F20110114_AADKJV cho_j_Page_012thm.jpg
fc3314e99ef5e4b0db5e592ded98ef6a
cb339193efa5d56028aef8098d9e38fc53ea65eb
115288 F20110114_AADKKL cho_j_Page_014.jp2
2a3dd309f3a38b21f36c195aed7d6855
c4f4f3f9128c617382a9c0e049eda63322e99db3
1573 F20110114_AADKJW cho_j_Page_100.txt
ab69d8ff7eac8f38e66bdeff0e2545f1
1dcd423db071fd1515a2b84f842aaf75f4e77be0
65542 F20110114_AADKLA cho_j_Page_066.QC.jpg
4389655db81c7023eb2b163eea866ab8
1b18ddb0ecd50fd53dca5c6cbc6b15d22c12a6e7
701470 F20110114_AADKJX cho_j_Page_052.jp2
a81b2c45495f2ee1df5cca235fc7f775
6c9f59a18ca6a773774d198ff96f2586cf39cbb3
34114 F20110114_AADKLB cho_j_Page_068.pro
2482a519813caf990e1d5550f98e7c19
caf9b71fa8c6f47ad6ead6afe980651f56193b7c
23631 F20110114_AADKKM cho_j_Page_038thm.jpg
9cea3ffdd5472b898106c888827f3035
b3ac5dbd17f2641d5fb906c70fc1a5abee653964
1051963 F20110114_AADKJY cho_j_Page_085.jp2
a365c11205287ec1724146aa9ad2092c
0f933815b52646ad8b7686ed7f2cb703cfb99e68
45814 F20110114_AADKLC cho_j_Page_099thm.jpg
e60276d72fcfe9d4f67127e5fecaa0b0
f76b87c4a3eebbeb82bb8b03ab0224e2a48d998a
F20110114_AADKKN cho_j_Page_046.tif
7446630215ab2926bd057f980b530be1
ce9e3a3dc2e3300fe6c07293136e65ac02e5e0aa
877358 F20110114_AADKJZ cho_j_Page_091.jp2
fc2f5bf8dcd324eb1743ed21c8950ea3
017b8d7f9fde6a9a5f2d183c1719f71db1fe807d
91318 F20110114_AADKLD cho_j_Page_083.QC.jpg
db4ed48fef60c5b17e0202458c881128
5ae311af15956b162591b8bf3d2738416423efbf
10772 F20110114_AADKKO cho_j_Page_130thm.jpg
dc088344a510b97781bd902fa2fa4eeb
2292c6d5238a0bfb5dd2b7683420175c725e15b7
77860 F20110114_AADKLE cho_j_Page_037.QC.jpg
f94437d50c998dbf8922f11bc1763b6b
972a7611fae8aa2871fb357ba003c9d1a5095e39
848730 F20110114_AADKKP cho_j_Page_100.jp2
713b0024a5c2707232f32eac917a0305
f781f4cf48ebda378bb9c76dd8b92a5dc433af19
894149 F20110114_AADKKQ cho_j_Page_061.jp2
44bda179cee50166a875c56b4b62b9a9
ffe3ba64e0eb3b719b4af7e59df7b1b7030f0671
85830 F20110114_AADKLF cho_j_Page_014.QC.jpg
8f4eafd2d6850404f9f1a12c2c21a5f2
ac6c12b4f514547034c2f244e612f5d482c6a6d8
1313 F20110114_AADKKR cho_j_Page_094.txt
ea9c42ccf9e298d3f0ba10e6005d303f
7aef8835395174267fb6694cb274173691a5c276
60880 F20110114_AADKLG cho_j_Page_010.pro
9fa1579a818f186098c364504727f332
3c68927685ab9d8960d66f3cbc6ed06c70c4a4b1
42076 F20110114_AADKKS cho_j_Page_050.pro
aab8b9d92fb59ca7828dd5bdf392bbc3
545d31d50398111cd05ddb64aad09c4cc19652f0
70493 F20110114_AADKLH cho_j_Page_105.QC.jpg
4687e47cd2839a0fcdfbb6d58482c85f
0430ec9750da89300086f44f487e2c0fc5efb54f
45330 F20110114_AADKKT cho_j_Page_092.pro
362c577fef821ee0df5b7b33f586c8db
00e8b55d9e8edaf6554008fec42cd9690f7ffa59
F20110114_AADKLI cho_j_Page_125.tif
5ea9093cf706f1411673d86460d44592
5c15a176a386b6d0da530765cf9b7962cac0f379
269543 F20110114_AADKKU cho_j_Page_125.jpg
bd11cd39d1da38084ee1906026f90123
1df758c984d42bd182bec9b6705d46410fee50f3
48292 F20110114_AADKLJ cho_j_Page_087thm.jpg
d7fa18252157165b0dd0468edb3aa715
f43f70508c40d3ca5267464afdc884adf678f578
F20110114_AADKKV cho_j_Page_012.tif
05051d2f1c7586460195b4eddf99a4b8
54f96f455d059f828bb22f1f4a8da5f29776c0cb
1889 F20110114_AADKLK cho_j_Page_066.txt
e592f0d06a326dab9e170a986b43141c
6574fda5c5bad4c500713d4197528eb929463290
F20110114_AADKKW cho_j_Page_021.tif
04342770c1fe69fb5d34107d4a9152e1
6bd978e3cc723426d493305233f4f85c03e46924
20704 F20110114_AADKLL cho_j_Page_069thm.jpg
84c71a5716d35a3d6658f8dc3efa4a3b
35a5909fd98b93af6793665422f22de954e94075
102082 F20110114_AADKKX cho_j_Page_017.jp2
0ce9d46e7bcef99cf99bc2f8ca7cf30b
2633dea715dc524ac24edbaa08cffed48c985356
73248 F20110114_AADKMA cho_j_Page_053.jp2
4ca426a63b11601e9e5e85a241624240
083aac95f172e993941c2b1514bc20ca806378db
76367 F20110114_AADKLM cho_j_Page_112.QC.jpg
a1d0a805f02190f0885ef7902e96379a
ce8ba28443f1e0b5b3fcd40ea884a859f381b4f1
213 F20110114_AADKKY cho_j_Page_003.txt
d449d1eac55f4ff2d814d5445d8592ae
4d6359a13205f66c901b58527a8d8d0424b3c197
F20110114_AADKMB cho_j_Page_121.tif
e5c4d0cbad13193a6b7c3844c268c8dc
6fa70a91baaf9c3b13a28bccbc1c7e832345c355
87557 F20110114_AADKKZ cho_j_Page_032.QC.jpg
a16ed4045d202c649d89c22f2d008feb
954a8d3e7aa31a01f625292a79c1423ee2c718a9
29722 F20110114_AADKMC cho_j_Page_082thm.jpg
98302544f56f69b9b0eb024f62ab8a1a
0e86498464bf4fc020a2d1789b4ca68a93e4165e
678 F20110114_AADKLN cho_j_Page_019.txt
4ff64a251dee41e7b109de2f232a6466
0710208fd2df3e73005db009cd695390d2e0ef90
22697 F20110114_AADKMD cho_j_Page_030thm.jpg
3d6aabbc3111e184b896d19e8da742c6
31a7e90173f6dfca7fd16a3c9135ea90a24f2ff0
23643 F20110114_AADKLO cho_j_Page_001.jp2
6c6f5777ba7293a84d83a7bf2f87d755
275bd82f6d553ec3db5d58a7f9e53811a04d0148
31572 F20110114_AADKME cho_j_Page_012.QC.jpg
be2fcaa5fcd853bfd2d0afdba2e3309a
fa224b389ba0da400e91b73f89a31188e923278c
204330 F20110114_AADKLP cho_j_Page_108.jpg
ebe7c219d394470421c2fa0cfca56a22
f9b127205f5c7b17015472a8756dbb727de6aedd
7885 F20110114_AADKMF cho_j_Page_001.pro
a7c2c4305df725c7011a492c8b36d93a
fd2392c8bbad5be1702a6071f3b33099e47e7527
24399 F20110114_AADKLQ cho_j_Page_055thm.jpg
8905be00d21136f38f1b9dfd4c952824
311a408e7cbe832eaf21c77ce2a73fe8533df7c8
135307 F20110114_AADKMG cho_j_Page_059.jpg
79136e20b271c996dab467b32ce4bfa2
dbfbffe4e06b55826166bb71e9a095a215ed78bb
213787 F20110114_AADKLR cho_j_Page_087.jpg
c89a8e81c69e8df557bd304d07e56913
45c07c3743850b3995cb658d13f42a5a498e48f5
56645 F20110114_AADKMH cho_j_Page_068.QC.jpg
09034dc0d369f1ebfc085214369fb114
781b84a2530ef4e8b9ae50e95252eac61ce95790
17713 F20110114_AADKLS cho_j_Page_059thm.jpg
60c44feda8486e0bae7bdceb06479146
5dc4852d1f51e765f10beadd206be62a1349dfed
194442 F20110114_AADKMI cho_j_Page_041.jpg
8369873f866b43efdd3248409977bb09
9539495a4ee5b930ecf4a01f54eb4349eb95fbde
50012 F20110114_AADKLT cho_j_Page_018.pro
3d060f620c0418d73dce9fee724a7401
5665c76985fd5b07a1d25e5c46841dd222b360ed
1844 F20110114_AADKMJ cho_j_Page_121.txt
336d5b844ff2ed6cfd59d05bcfd0dcc4
2957d40cc648c8aa38f41bbc2414cc1661863ffe
F20110114_AADKLU cho_j_Page_044.tif
dea4aea0ba1362f0ec7614df664e7dd9
993a8bbfa085328397314d68dac5a50c11f6b660


Permanent Link: http://ufdc.ufl.edu/UFE0008180/00001

Material Information

Title: Multiple Modeling and Control of Nonlinear Systems with Self-Organizing Maps
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0008180:00001

Permanent Link: http://ufdc.ufl.edu/UFE0008180/00001

Material Information

Title: Multiple Modeling and Control of Nonlinear Systems with Self-Organizing Maps
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0008180:00001


This item has the following downloads:


Full Text












MULTIPLE MODELING AND CONTROL OF NONLINEAR SYSTEMS
WITH SELF-ORGANIZING MAPS














By

JEONGHO CHO


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

JEONGHO CHO

































This document is dedicated to my late father My inspiration and the one who gave me
every opportunity to realize my dreams.















ACKNOWLEDGMENTS

Being at CNEL has been not only a wonderful academic experience but also a

unique opportunity to meet many colleagues who helped me immensely along the way.

It is a privilege to me to acknowledge the unconditional support of my supervisor,

Dr. Jose C. Principe, who has been a mentor during my years as a graduate student. His

advice, wisdom, and many invaluable lessons in life have made this dissertation possible.

I sincerely appreciate the help offered by the members of my academic committee,

Dr. John G. Harris, Dr. Michael C. Nechyba, and Dr. Loc Vu-Quoc. Their cooperation

and suggestions have considerable improved the quality of my dissertation. I would also

like to express my gratitude to Dr. Mark A. Motter and Dr. Deniz Erdogmus for their

help and providing many valuable comments during the past years of my stay in the

CNEL.

My deepest recognition goes to my beloved parents, especially to my late father

who helped me in any imaginable way to achieve my objectives and fulfill my dreams.

They have been an inexhaustible source of love and inspiration all my life.

My most special thanks go to my wife, Joonhee, for her patience, understanding

and encouragement, without which it would have been impossible to complete this

dissertation. Finally, I thank my specially beloved son, Minsuh, hoping that the effort of

these years may offer him a more plentiful life in the years to come.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ..................................... ........ .......................................... vii

LIST OF FIGURES ................................................ viii

A B S T R A C T ...................................................................................................... ............ x i

CHAPTER

1 IN TR O D U C T IO N ........ .. ......................................... ..........................................1.

1 .1 M o tiv atio n s ..................................................................................................... .
1.2 Review of Literature .................................................................. 2
1.3 Objectives and Author's Contribution.....................................................5...
1 .4 O u tlin e .......................................................... ................................................ 6

2 SYSTEM IDENTIFICATION VIA MULTIPLE MODELS .................................... 8

2 .1 L ocal D ynam ic M odeling ............................................................ ..................... 8
2.2 SO M -B ased L ocal M odeling ........................................................... ................ 10
2.2.1 R reconstruction of State-Space..................................................... 11
2.2.1.1 D elay reconstruction .................................................... 11
2.2.1.2 Estimating an embedding dimension ..........................................12
2.2.2 The Self-O organizing M ap ...................................................... ................ 14
2.2.2.1 C om petitive process .................................................... ............... 15
2.2.2.2 C cooperative process .................................................... ............... 16
2 .2 .2 .3 A daptive process ......................................................... .............. 17
2.2.3 Modeling Methodology Based on the SOM......................................... 17
2.3 Input-Output Representation of System s......................................... ............... 19
2 .3 .1 C classical A approach ................................................................... ............... 2 0
2.3.2 Series-Parallel and Parallel M odels....................................... ................ 21
2.3.3 SOM-based Multiple ARX Models.......................................................23
2.3.3.1 Selection of operating regions with a SOM ...............................23
2.3.3.2 M odel develop ent procedure ............................. ..................... 27

3 MULTIPLE MODEL BASED CONTROL ..........................................................29



v










3.1 D iscrete-Tim e C control System ........................................................ ................ 31
3.2 Inverse Control via Backpropagation Through Model....................................32
3.3 M multiple Inverse C ontrol........................................ ....................... ................ 34
3 .4 M multiple P ID C control ........................................... ......................... ................ 37

4 MULTIPLE QUASI-SLIDING MODE CONTROL ...........................................42

4.1 Introduction to Variable Structure Systems.....................................................42
4.1.1 Sliding H yperplane D esign.................................................... ................ 44
4.1.2 Sliding Mode Control Law Design ................ ...................................45
4.2 Sliding Mode Control in Sampled-Data Systems...........................................48
4 .2 .1 Q u asi-Sliding M ode................. ................................................. ............... 4 9
4.2.2 Quasi-Sliding Mode Control Using Multiple Models................................52
4.3 Analysis of Multiple Quasi-Sliding Mode Control with an Imperfect Sensor .....54

5 C A SE ST U D IE S ............... .. .................. .................. .............. ......... ... ............ 57

5.1 C controlled C haotic System s ............................................................ ................ 57
5.1.1 T he L orenz Sy stem ....................................... ...................... ............... 58
5.1.2 The D uffing O scillator .......................................................... ................ 66
5.2 Nonlinear Discrete-Time Systems....................................................72
5.2.1 A First-order Plant ............................. .............. ................................. 72
5.2.2 A Laboratory-scale Liquid-level Plant..................................................78
5 .3 F lig h t V eh icles ...................................................................................................... 8 6
5.3.1 M issile D ynam ics .. .. .... ........ ........ ............................................. 86
5.3.2 L oF L Y T E U A V .......................................... ........................ ................ 92

6 CONCLUSIONS AND FUTURE WORK........................................101

6 .1 S u m m a ry ............................................................................................................. 1 0 1
6 .2 F u tu re W ork ........................................................................................................ 10 3
6.3 C including R em arks ................. ............................................................. 104

APPENDIX

QSMC FOR MIMO SYSTEM .............................................................................106

LIST O F R EFEREN CE S .. .................................................................... ............... 109

BIOGRAPHICAL SKETCH ...............................................................................1...... 18















LIST OF TABLES


Table page

5-1 Lipschitz index of the controlled Lorenz system for determining an embedding
d im e n sio n ............................................................................................................... 5 9

5-2 Comparison of modeling performance for the controlled Lorenz system. .............62

5-3 Comparison of modeling performance for the controlled Duffing oscillator. .........67

5-4 Comparison of tracking performance for 3 different control task : Settling time
and NRAIS-SSE ....................... ........ ............... 70

5-5 Lipschitz index of a laboratory-scale liquid-level plant for determining an
em bedding dim pension. .............. .............. ............................................ 79

5-6 Comparison of modeling performance for the liquid-level plant..........................80

5-7 Comparison of control performance for the liquid-level plant in noise-free
en v iro n m e n t.............................................................................................................. 8 3

5-8 Comparison of control performance for the liquid-level plant in the presence of
sensor noise: standard deviation of noise is 4.5e-2. ............................ ................ 83

5-9 Comparison of modeling performance for the missile system...............................87

5-10 Comparison of controller performance for the missile system in the presence of
noise having the standard deviation of 4.6e-2..................................... ................ 90

5-11 Comparison of modeling performance for the lateral motion (p and r ) of the
LoFLYTE UAV. ................. ............. .. ......... .......................... 95















LIST OF FIGURES


Figure page

2-1 Local modeling scheme on the basis of a SOM ................................. ................ 10

2-2 Two data points that are close in (o but distant in (2 ...................... ...............13

2-3 K ohonen's Self-organizing M ap ......................................................... ................ 15

2-4 Nonlinear dynamic m odel configuration ............... .............. ..................... 20

2-5 A series-parallel model (left) and A parallel model (right).................................22

2-6 Configuration of local linear modeling based on a SOM ............... ..................... 27

3-1 Classical discrete-tim e control system ................................................ ................ 31

3-2 Modeling and control scheme using the TDNN: (a) TDNN modeling of a plant
(b) An inverse controller via Backpropagation through (Plant) Model ................34

3-3 Proposed SOM -based inverse control scheme ................................... ................ 36

3-4 P ID controlled sy stem ........................................................................... ................ 38

3-5 Overall schematic diagram of the nonlinear PID closed loop control mechanism
u sing m multiple controllers ........................................ ........................ ................ 39

3-6 Block diagram of PID controller for a SISO plant model...................................40

4-1 Phase plane plot of a continuous-time second-order variable structure system.......46

4-2 Discrete-time system response with sliding mode control..................................49

5-1 The uncontrolled Lorenz system: phase-space trajectory and time-series ............59

5-2 Generalization error v.s. Number of PEs (left) Learning curve (right).................60

5-3 Identification of the controlled Lorenz system by multiple models......................61

5-4 Tracking a fixed point reference signal by MIC: (a) yd =0 (b) yd =8 8 .............63









5-5 Comparison of control performance varying the number of inverse controllers
based on m multiple m odels ........................................ ........................ ................ 64

5-6 Comparison of tracking performance by multiple model based controllers (MIC,
MPIDC, MQSMC) and global inverse controllers (IC-ARX, TDNNC). ..............65

5-7 The uncontrolled Duffing oscillator: phase-space trajectory and time-series..........66

5-8 Lipschitz index (left) for the determination of optimal number of inputs and
outputs and Generalization error v.s. Number of PEs (right)...............................67

5-9 Identification of the controlled Duffing oscillator by TDNN (left) and multiple-
m o d els (rig h t) ........................................................................................................... 6 8

5-10 Performance comparison on trajectory tracking by TDNNC, PID-ARX, and
MPIDC when the poles of the closed-loop response are place at (a) 0.9 (b)
0.5 0.5i (c) 0.25 0.25i (d) 0.05 0.05i ........................................... ................ 69

5-11 Control performance by TDNNC (left) and MPIDC (right) ..............................71

5-12 Parameter selection to design multiple models .................................. ................ 72

5-13 Modeling performance using 64 multiple models for a nonlinear first
o rd e r p la n t. ............................................................................................................... 7 3

5-14 Responses for parameter selection to design QSMC by varying
(a) rT and (b) qT ........................................................................................... 74

5-15 Comparison of tracking performance using a global controller and multiple model
based controllers in the absence of sensor noise. The figure (right) is an
enlargement of the figure (left) between 34 and 52 iterations. ..............................75

5-16 Performance of square-wave tracking in the absence of noise by the MQSMC ...... 76

5-17 Comparison of performance against noise among TDNNC, MIC, MPIDC and
M Q S M C ............................................................................................................... .. 7 7

5-18 Sinusoidal and arbitrary signal tracking by the MQSMC:(a) in the absence of
sensor noise (b) in the presence of sensor noise, SNR = 20dB. ............................ 78

5-19 Modeling performance using multiple models for a liquid-level plant.................79

5-20 Typical input-output characteristic of the second-order liquid-level plant.............. 81

5-21 Square-wave tracking performance of the liquid-level plant varying the sliding
surface and the noise level by the M QSM C........................................ ................ 81









5-22 Control of the liquid-level plant by the MQSMC varying the number of
controllers: (a) M 1 (b)M= 16 (c)M= 36 (d)M= 144..................................82

5-23 Control of the liquid level system with measurement noise by the MQSMC
with (a)M= 1 (b)M= 16 (c)M= 144 and (d) the TDNNC. ..............................84

5-24 Tracking an oscillatory reference signal of the liquid-level plant by the TDNNC
(left) and the MQSMC (right) in the presence of sensor noise..............................85

5-25 Performance assessment on a trajectory tracking under noisy environment. ..........85

5-26 Modeling performance using multiple models for the missile dynamics. .............87

5-27 Tracking various set-point reference signal by (a) TDNNC (b) MIC (c) MPIDC
and (d) MQSMC in the absence of measurement noise ...................................... 88

5-28 Tracking various set-point reference signal by (a) TDNNC (b) MIC (c) MPIDC
and (d) MQSMC under the presence of measurement noise whose standard
d ev iatio n is 4 .6 e-2 ................................................................................................... 8 9

5-29 Trajectory tracking by TDNNC (left) and MQSMC (right) in the presence of
noise w hose standard deviation is 2.3e-1 ........................................... ................ 90

5-30 Set-point tracking behavior by the TDNNC (left) and the MQSMC (right) under
param eter variations. ............. ................. ............................. ...............9 1

5-31 General description of aircraft (left) and LoFLYTE testbed UAV (right) .............92

5-32 Control inputs (3, 3,) used to generate data samples for training the networks....94

5-33 M odeling of a roll-rate using multiple models.................................... ................ 95

5-34 Comparison for controlling roll-rate and yaw-rate to track the set point in the
absence of noise by (a) TDNNC (b) MIC (c) MQSMC ...................................... 98

5-35 Comparison for controlling roll-rate and yaw-rate to track the set point in the
presence of noise by (a) TDNNC (b) MIC (c) MQSMC. .................................... 99

5-36 Performance of controlling roll-rate and yaw-rate to track an arbitrary trajectory
with measurement noise (SNR = 20) by (a) TDNNC (b) MIC (c) MQSMC......... 100
















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

MULTIPLE MODELING AND CONTROL OF NONLINEAR SYSTEMS
WITH SELF-ORGANIZING MAPS

By

Jeongho Cho

December 2004

Chair: Jose C. Principe
Major Department: Electrical and Computer Engineering

This dissertation is concerned with the development and analysis of a nonlinear

approach to modeling and control of nonlinear complex systems. In particular, the

problem of designing a mathematical model of a nonlinear plant using only observed data

is considered.

For the identification of the plants, the concept of multiple models with switching

is employed in order to simplify both the modeling and the controller design since a

single controller may sometimes have difficulty meeting the design specifications in case

the dynamics vary considerably over the operating region. For this reason, a Self-

Organizing Map (SOM) is utilized to divide the operating region into local regions as a

modeling infrastructure to construct local models. The SOM selects the local operating

region relying on the embedded output, and the local model is built by the embedded

output as well as the embedded control input data samples which are spaced in the local

area.









Based on the identified multiple models, the problem of designing controllers is

discussed. Each local linear model is associated with a linear controller, which is easy to

design. Switching of the controllers is done synchronously with the active local linear

model that tracks different operating conditions. The effectiveness of the proposed

approach is shown through experiments for modeling complex nonlinear plants such as

chaotic systems, nonlinear discrete time systems and flight vehicles. Its comparison with

neural networks-based alternatives, Time Delay Neural Network (TDNN), shows clear

advantages of local modeling and control in terms of performance.














CHAPTER 1
INTRODUCTION

1.1 Motivations

The identification of nonlinear dynamical systems has received considerable

attention since it is an indispensable step towards analysis, simulation, prediction,

monitoring, diagnosis, and controller design for nonlinear systems [21,65,67]. In

particular, the problem of designing a mathematical model of a nonlinear plant using only

observed data has attracted much interest, both from an academic and an industrial point

of view. During the past few years, neural networks as a global model have been

suggested for nonlinear dynamical black-box modeling and successfully applied to the

prediction and modeling of nonlinear processes [11,46,65,73].

Global models, however, have shown some difficulties in cases when the

dynamical system characteristics vary considerably over the operating regime, effectively

bringing the issue of time varying parameters (or nonlinearity) into the design. On the

other hand, local modeling derives a model based on neighboring samples in the

operating space to characterize some operating point or similar feature [23,85]. If a

function f to be modeled is complicated, there is no guarantee that any given global

representation will approximate f equally across all space. Moreover, nonlinear models

are too complex to be used for controller design [70]. Thus, nonlinear control methods

cannot serve all needs of real industrial control problems.

In this case, the dependence on representation can be reduced using local

approximation where the domain off is divided into local regions and a separate model is









used for each region [3,13,24]. In a number of local modeling applications, a Self-

Organizing Map (SOM) has been utilized to divide the operating regions into local

regions [30,60,74,97]. The SOM is particularly appropriate for switching, because it

converts complex, nonlinear statistical relationships of high-dimensional data into simple

geometric relationships that preserve the topology in the feature space [44]. Thus the role

of the SOM is to discover patterns in the high dimensional state space and divide that

space into a set of regions represented by the weights of each Processing Element (PE).

Linear models and associated techniques for linear control design are typically used

to control the plant under certain specific operating conditions. This type of control is

only valid in a small region around the operating point. For that reason, the concept of

multiple models with switching, according to a change in dynamics, has been an area of

interest in control theory in order to simplify both modeling and controller design

[62,66]. The motivation for this research, therefore, is to explore control strategy using

SOM-based multiple models for nonautonomous and nonlinear systems.

1.2 Review of Literature

There are many examples in the literature in which the local modeling paradigm

has been successfully applied for the modeling of nonlinear autonomous and

nonautonomous systems. Farmer and Sidorowich [23] have shown that local linear

models, despite their simplicity, provide an effective and accurate approximation of

chaotic dynamical systems. Jacobs et al. [38] have proposed the mixtures of expert

models that are composed of several different expert networks and a gating network that

localizes the experts. They showed that a simple model can be built by dividing a vowel

discrimination task into appropriate subtasks. Bottou and Vapnik [3] have proposed using

local learning algorithms instead of training a complex system with all data samples, and









demonstrated that a set of subsystems trained with a subset of data can improve the

performance for an optical character recognition problem. The neural-gas architecture

proposed by Martinetz et al. [53] is similar to the SOM in that the competitive network

divides the input space in a set of smaller regions and then local linear models are created

by a LMS-like rule. They showed that the neural-gas network outperforms MLPs and

RBF networks for time series prediction. The same group [77] used a SOM for the

control of a robotic arm. Murray-Smith et al. [61] similarly have extended RBF networks

where each local model is a linear function of the input and exhibited great success in

control problems. Principe and Wang [74] have successfully modeled a chaotic system

with a SOM-based local linear modeling method. Vesanto [94] and Moshou and Ramon

[60] proposed a scheme that essentially followed local linear modeling based on SOM

topology for nonautonomous system. Under some conditions, it has been shown that

multiple models can uniformly approximate any system on a closed subset of the state

space provided a sufficient number of local models are given.

Generally, the control using multiple models is categorized by two approaches: a

global model-based control using local models and a multiple model-based control with

switching. Global controller design with the aid of multiple linear models has been

extensively reported in the literature. Gain scheduling has been perhaps the most

common systematic approach to control nonlinear systems in practice due to its simple

design and tuning [47,68,79]. The multiple model adaptive control approach differs from

gain scheduling mainly in the use of an estimator-based scheduling algorithm used to

weight the local controllers. Murray-Smith and Hunt [61] utilized an extended RBF

network where each local model is a linear function of the input and reported great









success for control problems. The overall controller designed is based on the local models

and a validity function to guarantee smooth interpolation. Similarly, Foss et al. [24] and

Gawthrop and Ronco [29] employed model predictive controllers and self-tuning

predictive controllers, respectively, using multiple models. Palizban et al. [70] attempted

to control nonlinear systems with the linear quadratic optimal control technique using

multiple linear models and provided the stability condition for the closed loop system.

Ishigame et al. [37] proposed the sliding mode control scheme based on fuzzy modeling

composed of a weighted average of linear systems to stabilize an electric power system.

In contrast, Narendra et al. [66] proposed the multiple model approach in the

context of adaptive control with switching where local model performance indices have

been used to select the local controller. Subsequently, Narendra and Balakrishnan [64]

proposed different switching and tuning schemes for adaptive control that combine fixed

and adaptive models yielding a fast and accurate response. Principe et al. [75] proposed a

SOM-based local linear modeling strategy and predictive multiple model switching

controller to control a wind tunnel and showed improved performance with decreased

control effort over both the existing controller and an expert human-in-the-loop control.

Later, Narendra and Xiang [63] proved that the adaptive control using multiple models is

globally stable and that the tracking error converges to zero in the deterministic case.

Diao and Passino [20] applied multiple model based adaptive schemes to the fault

tolerant engine control problem. A linear robust adaptive controller and multiple

nonlinear neural network based adaptive controllers were exploited by Chen and

Narendra [11]. Thampi et al. [89,90] have also shown the applicability of the multiple

model approach based on the SOM for flight control.









1.3 Objectives and Author's Contribution

This dissertation is concerned with modeling and control of nonlinear

nonautonomous dynamical systems. The objective of this dissertation is to investigate if

it is possible to obtain a better result in extending the formulation of the control problem

from using just one global model to using several internal models. Thus a multiple

modeling approach is presented and techniques to design controllers based on these

model structures are developed.

The main contributions made by the author with respect to the modeling and

control of nonlinear systems include:

Firstly, an extended version of the SOM-based local modeling scheme for

nonautonomous and nonlinear plants is developed for more general representation of the

underlying dynamical systems and better approximation solely based on input-output

measurements of the plant. Local linear models are derived through competition using the

SOM and they are derived from the data samples corresponding to each of the SOM's

PEs.

Secondly, we investigate several options regarding how to capture the dynamics in

the input-output joint space. It is shown that as the number of dependent variables is

increased SOM modeling may become increasingly difficult to model accurately due to

its memory based structure. Thus, the SOM is trained to position the local models in the

embedded output space. At any time instant, the model representing the plant dynamics is

chosen by the SOM depending on the history of the plant and then incorporated with the

previous control inputs.

Thirdly, the model structure is controller oriented since the dynamics are simpler

locally than globally such that it is easier to develop local models as well as controllers.









For instance, if the system phenomena or behavior changes smoothly with the operating

point, then a linear model (or controller) will always be sufficiently accurate locally

provided that the operating region is sufficiently small, even though the system may

contain complex nonlinearities when viewed globally. These local controllers can then be

"switched" as the system changes operating conditions. Hence, multiple control with

switching, such as an inverse controller and a PID controller using identified multiple

models, are examined.

Finally, in order to obtain a controller which preserves the good sensitivity to

external disturbances, a sliding mode controller (SMC) is employed using multiple

models. By doing so, one of the difficulties in designing a SMC (that requires the

complete knowledge of the plant to be controlled) can be removed. In addition, we

examine the effect by the modeling error due to the quantization of state space as well as

by measurement noise to the proposed multiple model based sliding mode control

performance. It is shown that the switching scheme does not create an issue to be

considered in order to guarantee BIBO stability of the overall system.

1.4 Outline

This dissertation is divided into six chapters. Chapter 2 gives a review of the local

dynamic modeling required to study further for nonautonomous systems. The SOM

employed as a modeling infrastructure to construct the local models is described briefly.

At the end, the proposed SOM-based multiple ARX modeling scheme is introduced for

nonlinear nonautonomous dynamic systems representation. Chapter 3 shows how to

design an inverse control and a PID control framework based on the designed multiple

linear models. A description of a global nonlinear TDNN trained by backpropagtion

through a model is also presented for comparisons regarding performance. Chapter 4









gives a brief description of Variable Structure Systems (VSS). Quasi-sliding mode

control strategy is proposed based on multiple models. In addition, analysis of multiple

quasi-sliding mode control structure with an imperfect sensor is discussed. In Chapter 5,

simulations are conducted assuming that both the plant is unknown and the only state

available for measurements is the plant output for two controlled chaotic systems, two

nonlinear discrete-time systems, one missile, and one Unmanned Aerial Vehicle (UAV).

Finally, Chapter 6 presents conclusions based on the preceding analysis and simulation

results and suggests further study.














CHAPTER 2
SYSTEM IDENTIFICATION VIA MULTIPLE MODELS

The idea of multiple modeling is to approximate a nonlinear system with a set of

relatively simple local models valid in certain operating regimes [39]. Because of the

complexity, uncertainty and nonlinearity of a large class of systems, we often cannot

derive appropriate models from first principles, and are not capable of deriving accurate

and complete equations for input-state-output representations of the systems. Hence we

need to resort to input-output data in order to derive the unknown nonlinear system model

[10,35]. The technique of multiple model networks is appealing for modeling complex

nonlinear systems due to its intrinsic simplicity [62,66].

2.1 Local Dynamic Modeling

We begin with a brief overview of a dynamical systems approach to input-output

modeling. When no physical knowledge of the system is available, we have to determine

a model from a finite number of measurements of the system's inputs and outputs. An

autonomous dynamical system's approach to "black-box" modeling based on Takens

Embedding theorem was first suggested by Casdagli [7]. The delay embedding offers the

possibility of accessing linear or nonlinear coupling between variables and is a

fundamental tool in nonlinear system identification. The use of delay variables in the

structure of these dynamical models is similar to that originally studied by Leontaritis and

Billings [49], and is common in linear time-series analysis and system identification [96].

When we are trying to understand an irregular sequence of measurements, an

immediate question is what kind of process generates such a series. Under the









deterministic assumption, irregularity can be autonomously generated by the nonlinearity

of the intrinsic dynamics. Let the possible state x of a system be represented by points in

a finite dimensional phase space, 9T This can be realized by a map of 9TP onto itself:

xk+1 Xk
: =f (2.1)
xk-P+2_ _xk-P+l _

The predictive mapping is the centerpiece of modeling since once determined, f can be

obtained from the predictive mapping f, : 9P -> 93 as

Xk+ = f (,k ) (2.2)

where Vx,k =[Xk, Xk-1, Xk-P+1]T. In addition, Singer et al. [85] derived the locally linear

prediction based on this relationship as

f(xk & V xk +b (2.3)

The vector and scalar quantities of d and b are estimated from the selected pairs

(x 1', i,,) in the least square sense, where j is the index of the data samples in the

operating regime, i.e., one model. To obtain a stable solution, more than P pairs must be

selected. In general, the above local model fitting is composed of two steps: a set of

nearby state searches over the signal history and model parameters which, when pieced

together, provide a global modeling of the dynamics in state space. The underlying

dynamics is then approximated as

f U f, (2.4)
=1,--..,N

where N is the number of operating regimes. Based on this approximation of an

autonomous system, local linear models have performed very well in comparative studies






10


on time series prediction problems and in most cases have generated more accurate

predictions than global methods [84,97]. Moreover, the nonlinear dynamical system can

be identified by local framework even in the presence of noise if enough data are

available to cover all of the state space since local regions are local averages of the data.

To make the local network less sensitive to noise and outliers, more than one neighbor

can be utilized in local modeling.

2.2 SOM-Based Local Modeling

The SOM is employed as a modeling infrastructure to construct the local models. It

provides a codebook representation of the plant dynamics and organizes the different

dynamic regimes in topological neighborhoods. Thus we can create a set of models that

are local to the data in the Voronoi tessellation created by the SOM. This local model

structure with the SOM is depicted in Figure 2-1.

uk Yk



Reconstruction of State-Space



Yk+ 1 -- Model l Switching
|IModel 2 Device
-: (SOM)

.Model N


Figure 2-1. Local modeling scheme on the basis of a SOM.









2.2.1 Reconstruction of State-Space

In many cases of practical interest it is not possible to measure the state variables of

a system directly. Instead, the measuring procedure yields some value yk = (k), when

the system is in states xk. Here, (p(-) is a measurement function which in general

depends on the state variables in a nonlinear way. The time evolution of the state of the

system results in a scalar time series x,, x, x3,.... In order to reconstruct the underlying

dynamics in phase space, delay embedding techniques are commonly used.

2.2.1.1 Delay reconstruction

Delay-coordinate embedding [41,91], a technique developed by the dynamics

community, is one way to help the input-output modeling; it allows one to reconstruct the

internal dynamics of a complicated nonlinear system from a single time series. That is,

one can often use delay-coordinate embedding to infer useful information about internal

(and immeasurable) states using only output information. The reconstruction produced by

delay-coordinate embedding is not, of course, completely equivalent to the internal

dynamics in all situations, or embedding would amount to a general solution to control

the theory's observer problem: how to identify all of the internal state variables of a

system and assume their values from the signals can be observed [87]. However, a single-

sensor reconstruction, if done properly, can still be extremely useful because its results

are guaranteed to be topologically (i.e., qualitatively) identical to the internal dynamics.

This means that conclusions drawn about the reconstructed dynamics are also true of the

internal dynamics of the system inside the black box. In order to reconstruct the

underlying dynamics in phase space, we begin with scalar observable, Yk of the state xk









of a deterministic dynamical system. Then typically we can reconstruct a copy of the

original system by considering blocks

V'y,k =[Q( (k)(k-kdy-I 4 (2.5)
= [Yk Yk- ''"Yk (dy-)T \t

of dy successive observations of
reconstruction is governed by two parameters, embedding dimension dy and embedding

delay r. Note that using dy = 1 merely returns the original time series; one-dimensional

embedding is equivalent to not embedding at all. Proper choice of dy and r is critical to

this type of phase-space reconstruction and must therefore be done wisely; only "correct"

values of these two parameters yield embeddings that are guaranteed by the Taken's

theorem [88] and subsequent work by Packard et al. [69] and Casdagli et al. [7] to be

topologically equivalent to the original (unobserved) phase-space dynamics.

2.2.1.2 Estimating an embedding dimension

There has been much work on determining the embedding dimensions of the time

series generated by autonomous dynamical systems in the absence of dynamical noise

[41,91,42,6]. The methods developed for estimating the minimum embedding dimensions

are grounded on Takens' embedding theorem [88] and most of them use the ideas of the

false nearest neighbors technique [42,6]. Later a number of works discussed theoretical

foundations of the delay embedding of the input-output time series [7,6]. This led to the

generalization of the existing method for the case of non-autonomous dynamical systems

[87,6,76].

He and Asada [35] proposed a strategy which is based directly on measurement

data and does not make any assumptions about the intended model architecture or










structure. It requires only that the process behavior can be described by a smooth

function, which is an assumption that must be made in black box nonlinear system

identification. An explanation of this strategy's central idea follows. In general case, the

task is to determine the relevant inputs of the function

Yk+1 1f( 1,(02,',(On) (2.6)

from a set of potential inputs (p1, ( o2 ... (, ( (o > n) that is given. If the function in (2.6)

is assumed to depend on only n 1 inputs although it actually depends on n inputs, the

data set may contain two (or more) points that are very close (in the extreme case they

can be identical) in the space spanned by the n -1 inputs but differ significantly in the

nth input. This situation is shown in Figure 2-2 for the case n = 2.


(,2

............................................., Y



YJ
....................................... j





Figure 2-2. Two data points that are close in (p, but distant in (P2

The two points i andj are close in the input space spanned by (,1 alone but they are

distant in the (1 (92- input space. Because these points are very close in the space

spanned by the n 1 inputs ((0p) it can be expected that the associated process outputs y,

and yj are also close (assuming that the function f(.) is smooth). If one (or several)


relevant inputs are missing then obviously y, and yj are expected to take totally









different values. In this case, it is possible to conclude that the n 1 inputs are not

sufficient. Thus, the nth input should be included and the investigation may begin again.

In [35] an index is defined based on so-called Lipschitz quotients, which is large if

one or several inputs are missing (the larger the quotients, the more inputs are missing)

and is small otherwise. Thus, using this Lipschitz index the correct embedding

dimensions can be detected at the point where the Lipschitz index ceases to decrease. The

Lipschitz quotients in the one-dimensional case are defined as


1, = for i, j 9L, i j (2.7)


where L is the number of samples in the data set. For the multidimensional case, the

Lipschitz quotients can be calculated by the straightforward extension of (2.7):


(Y1= ( fori, j \ 2L, ij (2.8)


where n is the number of input. The Lipschitz index, then, can be defined as the

maximum occurring Lipschitz quotient

I" = max (/;) (2.9)


As long as n is too small and thus not all relevant inputs are included, the Lipschitz

index will be large because situations as shown in Figure 2-2 will occur. As soon as all

relevant inputs are included, (2.9) stays relatively constant.

2.2.2 The Self-Organizing Map

The principal goal of the SOM is to transform an incoming signal pattern of

arbitrary dimension into a one or two-dimensional discrete map, and to perform this

transformation adaptively in a topologically ordered fashion [44]. Figure 2-3 shows









Kohonen's model of a two-dimensional SOM. Each PE in the lattice is fully connected to

all the source PEs in the input layer. This network represents a feedforward structure with

a single computational layer consisting of PEs arranged in rows and columns.



x1,k 0 O Winning PE







Input Layer Competition Layer


Figure 2-3. Kohonen's Self-organizing Map.

The algorithm responsible for the formation of the SOM proceeds first by

initializing the synaptic weights in the network. Once the network has been properly

initialized, there are three essential processes involved in the formation of the SOM:

competition, cooperation and adaptation. Descriptions of these processes follow.

2.2.2.1 Competitive process

Let m denote the dimension of the input space. Let an input vector selected

randomly from the input space be denoted by

Xk =[xl,k, 2,k, ***,Xk (2.10)

The synaptic weight vector of each PE in the network has the same dimension as in the

input space. Let the synaptic weight vector of PE i be denoted by

i, = [w ,,W2,,,---,W ,, ], i G N (2.11)

where Nis the total number of PEs in the network.









To find the best match of the input vector xk with the synaptic weight vectors w,,

compare the Euclidean distance between xk and v, and select the smallest one as

i = ii it i =1,---,N (2.12)


which sums up the essence of the competition process among the PEs. According to

(2.12), i is the subject of attention because we want the identity of PE i. The particular

PE i that satisfies this condition is called the best-matching unit or winning PE for the

input vector xk.

2.2.2.2 Cooperative process

The winning PE locates the center of a topological neighborhood of cooperating

PEs. A topological neighborhood can be defined by many methods. In particular, a PE

that is firing tends to excite the PEs in its immediate neighborhood more than those

farther away from it, which is intuitively satisfying. This means that after classification of

the input sample, the adaptation will be done not only for the winning PE but also for the

neighbors of the PE which gives the best response. Let A ,, denote the topological


neighborhood function centered on the winning PE i, then a typical choice of A, is



A -o, -rpo2 (2.13)
2rk


where r, ro represents the Euclidean distance in the output space between the ith PE

and the winning PE and rk is the effective width of the topological neighborhood. To

satisfy the requirement that the size of the topological neighborhood shrinks with time, let

the width of the topological neighborhood function decrease with time










ak =oexp k-), k=1,2,... (2.14)


where o- is the value of c at the initiation of the SOM algorithm, and o, is a time

constant. Thus, as time (i.e., the number of iterations) increases, the width decreases at an

exponential rate, and the topological neighborhood shrinks in a corresponding manner.

2.2.2.3 Adaptive process

The network can be trained with a simple Hebbian-like rule to train the weights of

the winning PE and its neighbors. The neighboring PEs can be trained in proportion to

their activity (Gaussian), or all neighbors within a certain distance can be trained equally.

The learning rule can be described as follows:

,+ = + k (2.15)
1 Wlk, O1lh '/ 11 i t'

Notice that both the learning rate, l7k, and neighborhood size, -k, are time dependent

and are typically annealed (from large to small) to provide the best performance with the

smallest training time.

2.2.3 Modeling Methodology Based on the SOM

In this architecture of local linear modeling, the SOM is trained to position the local

models in the embedded output space, fy,k =[YkYk 1,'" Yk-(dy -1),. The SOM

preserves topological relationships in the input space in such a way that neighboring

inputs are mapped to neighboring PEs in the map space. Then, when each PE is extended

with a local model it can actually learn the mapping yk+l = f(y,k) in a supervised way.

Each PE has an associated local model { 5i,, b, } in (2.3) that represents the approximation

of the local dynamics.









The local model weights { ,, b, } are computed directly from the desired signal

samples y, and the input samples by a least square fit within a Voronoi region centered

at the current winning PE chosen from f7y. The size of the data samples in the region

must be at least equal to the dy -dimensional basis vector. The design procedure for this

local model is as follows:

1. Apply training data to the SOM and find the winning PE corresponding to the input
V/Y such that we have winner-input pairs.

2. Use the least square fit to find the local linear model coefficients for the winning
PE, i, where desired output vector y, e ^3M as

y, = [, bo] ,Io for Vj e M (2.16)

where = [o bo ] is the sought linear model coefficients, Mis the size of inputs
involved in the winning PE i.

Specifically, the least-squares problem

Y = OX (2.17)

is solved for 0, where X e 9t (dM) is defined as a matrix that contains each input vector

associated with the winning PE, and Y e 91^ is defined as a vector that contains the

target outputs. It is well-known that although the least-squares solution obtained from

(2.17) is reasonably good when the noise level is low, the estimates tend to be biased for

higher levels of noise. Addition of a single sample to a cluster can radically change the

distances. Besides, the models will perform very well for that particular training set with

very low error because it has memorized the training examples but they may not perform

well with new data sets. Thus we make use of data samples from the winner as well as

the neighbors to create the local models in order to make them more robust as well as to









improve the generalization for the network. Also we take the data samples from the

neighbors in case less data than the dimension of the input are assigned in some Voronoi

region.

In testing, once the winning PE is determined we select the appropriate local model

from the list of associated models. Apply the local model to obtain the estimated output

Yk+1 = y Vy,k + b, (2.18)

2.3 Input-Output Representation of Systems

The temporal state evolution of an autonomous system is functionally dependent

only on the system state, but a nonautonomous system, such as considered in this work,

allows for an explicit dependence on an independent variable, the control input, in

addition to the system state. For an autonomous system, it is reasonable to assume that

the future behavior of the system can be predicted over some finite interval from a finite

number of observations of past outputs. In contrast, predictions of the behavior of a

nonautonomous system require consideration of not only the "internal" deterministic

dynamics (past outputs), but also of the "external" driving term (future input) [30,75,96].

System identification is a technique that permits building mathematical models of

dynamic systems based on input-output data (measurements). Its main purpose is to

identify a model of an unknown process in order to predict and gain insight into the

behavior of the process [39]. Real-life systems almost always show nonlinear dynamical

behavior. This behavior complicates the task of finding models that accurately describe

these systems. While in a large number of applications a linear model shows already

satisfactory results, there are numerous situations where linear models are not accurate

enough; especially when we deal with very complex systems or require very high









performance. Physical knowledge of the system can be a great aid in finding a nonlinear

model. However, this knowledge is not always available. In these cases we have to

determine a model from a finite number of measurements of the system's inputs and

outputs. This approach to nonlinear system modeling is often referred to as nonlinear

black-box identification. Usually, a nonlinear mapping is fitted from a number of delayed

inputs and outputs to the current output [94]. This results in a nonlinear input-output

model of the system.

2.3.1 Classical Approach

Some common classical approaches for nonlinear nonautonomous system modeling

are based on polynomials, e.g., Kolmogorov-Gabor polynomial models [71], Volterra

Series models [110], Hammerstein models [16,22,33], and so on, for the realization of the

nonlinear mapping.


21k 1 1 z1 z Yk



Nonlinear static approximator f(.)



Yk+l

Figure 2-4. Nonlinear dynamic model configuration.

Normally, a discrete-time nonlinear dynamic system can be described by a NARX

(Nonlinear Auto-Regressive with eXogenous input) model that is an extension of the

linear ARX model, and represents the system by a nonlinear mapping of past inputs and

output terms to future outputs, that is,

k+l =f(Yk, ,Yk yk- 1, k, *,U k- /+1) (2.19)









Here Yk e Y c 9'P is the output vector and u.k e Uc 9q' is the input vector. For

simplicity, we will set p = q = 1. Let the (dy + du) dimensional basis vector be

/k = [vy,k, Vu,k] = [Yk,"**Yk -dy+l, k, k-d+1 ] (2.20)

where W/k is in the set = Yy x U'd. Figure 2-4 shows the schematic diagram of NARX

model. Another nonlinear model is a NOE (Nonlinear Output Error) structure described

by

Yk+l = f(k,'-" k -dy+l, 1k k-du+1) (2.21)

where y is the output of the identification model f. A NARMAX (Nonlinear Auto-

Regressive Moving Average with eXogenous input) is the lagged version of the NARX

model and is represented by

Yk+l = f(Yk, "Yk- dy+ ,k,'**,k-du )+ek (2.22)

In the above, f can be replaced by neural networks, radial basis function networks or

fuzzy logic systems, which are other methods that have been developed for nonlinear

system identification [5,96]. Narendra and Parthasarathy [65] have compared NOE and

NARX, and as a result they have shown that NARX is better than NOE. In the neural

network community most identification schemes use the series-parallel model (NARX).

2.3.2 Series-Parallel and Parallel Models

A nonlinear dynamic model can be used in two configurations: a "series-parallel"

model and a "parallel" model. A series-parallel model predicts one or several steps into

the future on the basis of previous plant inputs and plant outputs and ensures that all the

signals are bounded if the plant is BIBO stable. Most published reports use the series-

parallel model because of its resulting stability. A requirement for using this model is that









the plant output is measured during the operation. In particular, in control engineering

applications the series-parallel model plays an important role, e.g., for the design of a

minimum variance or a predictive controller.


Plant, P(z) Plant, P(z)





Model, f(.) Model, f(.)



Yk+1 Yk+1

Figure 2-5. A series-parallel model (left) and A parallel model (right).

In contrast, a parallel model is required whenever the plant output cannot be

measured during operation. This is the case when a plant is to be simulated without

coupling to the real system, or when a sensor is to be replaced by a model. Also, for fault

detection and diagnosis the plant output may be compared with the simulated model

output in order to extract information from the residuals. Finally parallel model is very

useful when dealing with noisy systems since it avoids problems of bias caused by noise

on the real system output: If the identification model is to be used offline, the parallel

model is obviously more suitable. The parallel model, however, lacks theoretical

verification; hence, it is difficult to utilize its advantages.

The two configurations shown in Figure 2-5 can not only be distinguished for the

model operation phase but also during training. In this research, we follow the series-

parallel model.









2.3.3 SOM-based Multiple ARX Models

In the interest of modeling the local dynamics of a nonautonomous system in each

region, the local approximation method presented for autonomous systems can be

extended by letting xk Vk in (2.3), so that yk+1 f('k). Provided that necessary

smoothness conditions on f, : -> Y are satisfied, a Taylor series expansion can be

used around the operating point. The first-order approximation about the system's

equilibrium point produces N local predictive ARX models f, ---, f, of the plant

described by

dy-1 du-1
f(k) Za,yk j + bJUk ,i =1,...,N (2.23)
J=0 J=0

where a,,j and b,,j are the parameters of the ith model. Although higher order Taylor

approximations would improve accuracy, they are not very useful in practice because the

number of parameters in the model increases drastically with the expansion order.

Our proposed methodology is summarized as follows: first, the delayed version of

input-output joint space is decomposed into a set of operating regimes that are assumed to

cover the full operating space Next, for each operating regime we choose a simple linear

ARX model to capture the dynamics of the region. Consequently, a nonlinear

nonautonomous system is approximated by a concatenation of local linear models

f()W U f (y, V) (2.24)
i=1,.--,N

2.3.3.1 Selection of operating regions with a SOM

Building local mappings in the full operating space is a time and memory

consuming process, which led to the natural idea of quantizing the operating regimes and

building local mappings in positions given by prototype vectors obtained from running









the plant. For quantization of the operating regimes, the k-nearest-neighbor method is

effective but it disregards neighborhood relations, which may affect performance [53]. In

contrast, the SOM has the characteristic of being a local framework liable to limit the

interference phenomenon and to preserve the topology of the data using neighborhood

links between PEs. Neighboring PEs in the network compete with each other by means of

mutual lateral interactions, and develop adaptively into specific detectors of different

signal patterns [44]. The training algorithm is simple, robust to missing values, and it is

easy to visualize the map. These properties make SOM a prominent tool in data mining

[94].

In most of the papers discussing local linear models for system identification, the

SOM has been used with a first order expansion around each PE in the output space. The

SOM transforms an incoming signal pattern of arbitrary dimension into a one or two-

dimensional discrete map, and performs this transformation adaptively in a topologically

ordered fashion [44]. The results obtained so far with this methodology have been quite

promising. However, problems that need to be solved remain: first, efficiently

partitioning the operating regimes in high dimensional spaces is still a problem due to the

curse of dimensionality [30]; second, it may be hard to find a small number of variables

to characterize the operating regimes due to the possibly large number of local models;

third, all the methods have to be extended for nonautonomous regimes.

The previous work by Principe et al. [75] provided the starting point for the

proposed modeling architecture. The most important difference is how to capture the

dynamics in the input-output joint space, which is fundamental for identifying the

unknown system. Several options are possible, and we have been investigating them:









Firstly, we tried to find the local models by quantizing the input-output joint space

by embedding not only the outputs but also the control inputs using one SOM. This

modification is essential because the purpose is to characterize the system dynamics that

exist in the input-output joint space. However, we encountered some difficulties such as

normalization of the joint space and large dimensionality of the space involved (many

degrees of freedom and large dynamic range of parameters) [13].

Secondly, in order to reduce the approximation error with local models based on a

SOM, we utilized a counter-propagation network by quantizing the input-output joint

space and the desired signal space together [15]. Since the output at each PE is just the

average output for all of the feature vectors that map to that point local models might be

created for better approximation using the quantization error in the input space and the

average output. This is achieved by coupling each PE with a linear mapping in such a

way that a functional relationship can be established between each Voronoi region in the

input space (of the SOM) and the desired signal. However, this method required a much

larger map to make the estimation error in the desired output space smaller. Additionally,

when noise is added in the input of the SOM, the quantization error in the input Voronoi

region may be magnified by the local models.

As the number of dependent variables is increased, the process becomes

increasingly difficult to model accurately. This led us to think that a model that uses only

a few of the observed variables will be more accurate than a model that uses all the

observed variables. In this scheme, therefore, we let the SOM look at only the current

output and its past values to decide the winner, and create the models with the control

inputs.









Here we will pursue the last option for the following reasons. The competitive

learning rule works best for normalized inputs. The SOM algorithm uses the Euclidean

metric to measure distances between vectors. For example, if one variable has values in

the range of [-100,...,100] and another in the range of [-1,...,1] the former almost

completely dominates the map organization because of its greater impact on the measured

distances. Either, the measure of distance is weighted by the inverse of the scales or the

data must be normalized such that each component of the input vectors have unit variance

and zero means [8]. However, normalization loses information (the mean or the scale can

be important) and it can become meaningless if the data dynamic range (or mean)

changes over time. Therefore we cannot normalize the data (nor create the weighted

Euclidean metric) in this way since it is not always guaranteed that the mean and the

dynamics range of the data are available.

In addition, as the number of dependent variables is increased, SOM modeling

becomes increasingly difficult because it is basically a memory-based approach that does

not scale up well with the input dimension. This led us to think that a model that uses

only a few of the observed variables will be more accurate than a model that uses all of

the observed variables. When the SOM modeling is done in the output space, we let the

SOM look at only the current output and its past values to decide the winner which

represents the operating regime, and create the models with the control inputs as shown

in Figure 2-6. In so doing, normalization of the input space is not necessary since the

clustering is performed solely by the history of the output.






























Figure 2-6. Configuration of local linear modeling based on a SOM.

2.3.3.2 Model development procedure

After the operating regions are divided by the SOM the underlying dynamics f is

then approximated as f U 11 f where N is the number of operating regions. N local

predictive ARX models f1, fN of the plant are described by

dy du
f/( k) aJyk- +ZbJuk-J, i=1,...,N (2.25)
J=0 J=0

where a,,j and b,, are the parameters of the ith model. Then, when each PE of the SOM is

extended with a local model it can actually learn the mapping yk+l = f (Vy,k V.u,k) in a

supervised way. The development of local models is done by directly fitting the

quantized embedded output samples obtained from the SOM and corresponding

embedded control input samples that cover the whole range of operation of the plant.









Each PE has an associated local model { d,, b} which are computed directly from

the desired signal samples r,, and the input-output samples by a least square fit within a

Voronoi region centered at the current winning PE chosen from y,. The design

procedure for this local model is as follows:

1. Apply training data to the SOM and find the winning PE corresponding to the input
Vy such that we have winner-input pairs.

2. Use the least square fit to find the local linear model coefficients for the winning
PE, i, where desired output vector r, e 91M as


r,= b = a Y '] for Vj eM (2.26)


where [ao b o] is the sought linear model coefficients, M is the size of data
involved in the winning PE i.

3. In testing, once the winning PE is determined we select the appropriate local model
from the list of associated models. Apply the local model to obtain the estimated
output

vk+l y,k +bTk (2.27)

Our proposed modeling methodology is summarized as follows: first, the delayed

version of input-output joint space is decomposed into a set of operating regions that are

assumed to cover the full operating space. Next, for each operating region we choose a

simple linear ARX model to capture the dynamics of the region. Consequently, a

nonlinear nonautonomous system is approximated by a concatenation of local linear

models.














CHAPTER 3
MULTIPLE MODEL BASED CONTROL

Researchers have been interested in control of nonlinear systems for a very long

time. Progress in nonlinear control design, however, has been difficult because of the

intrinsic complexity of the problem [82]. In general, nonlinear control methods are

complex and can be applied only to a narrow class of systems. For example, methods

such as backstepping and feedback linearization can be applied to nonlinear systems with

some specific structure, but not to arbitrary nonlinear systems. Thus, nonlinear control

methods cannot serve all needs of real industrial control problems.

One way to approach the control of a nonlinear system in a wide range of

conditions is to linearize the model at a number of operating points, and then design one

linear feedback controller at each operating region [101]. These local controllers can then

be "switched" or "scheduled" as the system changes operating conditions. The use of

multiple models is not novel in control theory. Multiple Kalman filters were proposed in

the 1960s and 1970s by Magill [51] and Lainiotis [45] to improve the accuracy of the

state estimates in control problems. Fault detection and control in aircraft was proposed

by Maybeck and Pogoda [54], and in the subsequent years Maybeck and Stevens [55]

used the idea extensively in controlling aircraft systems. In all the above cases, no

switching is concerned, and only a linear combination of the control determined by the

different models is used to control the system.

The idea of switching between controllers has been most likely introduced for the

first time in the adaptive control literature by Martensson [52]. In the direct switching









schemes, the sequence in which the different controllers are to be tried is pre-determined.

The only determination that has to be made is when to switch from one controller to

another. It was soon realized that such architectures have very little practical utility. On

the other hand, the outputs of the multiple observers determine both when and to which

controller switching should occur in indirect switching schemes. Middleton et al. [56]

explicitly proposed the use of multiple models and switching to alleviate the problem of

stabilizing of the estimated model in indirect control, and further extended in [59] and

labeled the "Hysteresis switching algorithm". The objective in all the above efforts is to

attain stability in adaptive control with minimum past information.

A controller is often highly dependent on a plant model especially when the

controller has been designed out of the model. Hence, for those cases, the modeling error

would be a relevant criterion for controller selection. If the number of controllers is

bounded, the delay between the selection of the controller and its activation can be

neglected. Thus the selection of the controller according to the modeling error is feasible.

Narendra and Balakrishnan [64] were the first to propose this idea of using multiple

adaptive models and switching in order to improve the performance of an adaptive

system, while assuring stability. Although it has already been shown that the performance

of a system can be significantly improved using the multiple model adaptive control with

switching, applicability to highly complex systems with this approach has not been

investigated in details. Thus, in this chapter, multiple controller design methodologies are

introduced by extending the multiple model approach for more complex nonlinear

systems.









3.1 Discrete-Time Control System

The field of classical control theory concerns itself with the task of servo or

regulator control of linear analog plant. Design methods for both continuous-time linear

controllers and discrete-time linear controllers obtained by discretizing the plant are well

understood. Figure 3-1 shows a schematic diagram of a classical discrete-time control

system.

dk
rk -1 Controller uk Plant Yk+1
-o G(z) P(z)





Figure 3-1. Classical discrete-time control system.

The signal rk1 is the reference signal. We would like the plant output Yk+1 to track

it as closely as possible. To track the reference signal, the controller uses both rkj and

Yk to compute the plant control signal uk. Feedback of Yk is used to stabilize the plant,

and to ensure that the controller is both resilient in the face of external disturbances and

able to quickly reduce the output error to zero. Their only drawback is that they assume

precise knowledge of the plant dynamics. For this reason, a great deal of effort has been

expended to create accurate models of typical plants. As one improved way of modeling,

we proposed multiple models for better approximation of the plant in Chapter 2. This

scheme makes it easier to design the controller for the model which approximates internal

descriptions of the plant when given a finite number of external measurements without

any knowledge of the plant dynamics.









One reason to consider discrete-time systems is that it is well known that most

complex systems are controlled by computers which are discrete in nature and this

constitutes an obvious reason for dealing with multiple model based control. Another is

the fact that the presence of random noise can be dealt with more easily in the case of

discrete-time systems. Since most practical systems have to operate in the presence of

noise, the stability and performance of multiple model based control in such contexts has

to be well understood, if the theory is to find wide applications in practice.

3.2 Inverse Control via Backpropagation Through Model

In order to design a controller, we need to determine a plant model which should

capture the dynamics of the plant well enough that a controller designed to control the

plant model will also control the plant very well. Such a model might be derived from

physics by carefully analyzing the system and determining a set of partial differential

equations which explain its dynamics. Alternatively, the model might be a black-box

implementing some sort of universal transfer function. This function may be tuned by the

adjustment of its internal parameters to capture the dynamics of the system. For nonlinear

unknown systems, a NARX model of sufficient order is a universal dynamic system

approximator. Hence, we implement NARX neural network plant models for

performance comparisons with the proposed local linear models.

The conventional design methods for control systems involve constructing a

mathematical model of the system's dynamics and utilization of analytical techniques for

the derivation of a control law. Such mathematical models comprise sets of linear or

nonlinear differential/difference equations, which are usually derived with a degree of

simplification or approximation. The modeling of physical systems for feedback control

generally involves a balance between model accuracy and model simplicity [102]. Should









a representative mathematical model be difficult to obtain, due to uncertainty or sheer

complexity, conventional techniques prove to be less useful. Also, even though an

accurate model may be produced, the underlying nature of the model may make its

utilization using conventional control design difficult.

Neural networks, hence, have been used for different purposes in the context of

control due to its ability to learn an essential feature of unknown plants by mimic

[43,105,78]. Most of methods existing are based on inverse control. We will therefore

start by considering a neural network controller, specifically, TDNN since this is very

suitable to create both a model and a controller when only input-output measurements are

available.

Principally, the TDNN is an extended multilayer perception that allows us to

handle temporal patterns and the problems of time variant signals, i.e., signals that are

scaled and translated over time [40]. The idea that has been followed in the TDNN is

based on the invention of time delays, resulting in giving the individual PEs the ability to

store the history of their input signals. This way, the network as a whole can adapt not

only to a set of patterns, but also to a set of sequences of patterns. An advantage of the

TDNN is the relatively simple mathematical analysis and ability of training by

Backpropagation algorithm. Thus we compare the performance of the proposed control

systems with that of an inverse controller trained through the TDNN.

The algorithm is derived as follows: First we train a TDNN as a model, f, by

letting yk+1 = f(Yk 1, ,Uk, Uk-,I i- -) as shown in Figure 3-2(a). Then a TDNN

controller is designed based on the created TDNN model from which we obtain the

Jacobian of the plant. The controller is described by









1k = g(uk-1, Uk-k ,r+,y,k k-1,-,W) letting g be the function implemented by the

controller G(z). Here W is the weights of the TDNN model, F(z). The controller

parameters in the fixed control structure are adapted by an algorithm that ensures that the

desired performance level is maintained and the parameters are updated by back

propagating the error through the model as shown in Figure 3-2(b).




Embedding Embedding










3.3 Multiple Inverse Control

Now we discuss the control problem for the local linear model using an inverse

control framework [17]. The central advantage of such a framework is that an inverse

model can be used directly to build a feed-forward controller. When given a model

Yk+1 = f (Yk,Yk-1,Uk,Uk- ), the control network is brought on line and the control signal

is calculated at each instant of time by setting the output value Yk+1 at instance k +1

equal to the desired value rk+1 as 1uk -f 1 (Yk, Yk 1, rk+1, k 1) while trained off line as in

classical inverse control approach. Thus, for the desired behavior, the controller simply

asks the model to predict the action needed.









As stated before, our principal objective is to determine a control input, uk, which

will result in the output, yk+l of the plant tracking with sufficient accuracy a specified

sequence, rk+1. The system identification block has N predictive models denoted by


{f }N1, in parallel. Corresponding to each model /, a controller g, is designed such that

g, achieves the control objective for f Therefore, at every instant one of the models is

selected and the corresponding controller is used to control the actual plant. In order to

control a plant, consider the control problem where the dimension of the input is equal to

that of the output, that is, p = q. From (2.23), because p = q, and under the assumption

that b, is invertible, the control law of an inverse controller for the model, f can be

directly calculated as

S dy-1 du-
uk = oo rk+ -1~Zao -y --Zbo Uk- (3.1)
J ]=0 J =1

Therefore, at time instance k, the control uk can be obtained, if the future target of Yk,

rkl is known. Therefore, the set of local linear models simplifies the control design for a

nonlinear plant. So instead of a global neuro-controller as in other adaptive control

schemes [12,18,31,58], here we can function with a group of linear controllers associated

with each identified model, thus taking care of the system over the whole operating

region.

One advantage of this scheme is its simplicity and fast convergence to get the

desired response. Another advantage is that the dynamic space is decomposed in the

appropriate switching among very simple linear models, which leads to accurate

modeling and controls. On the other hand, creating a set of models by embedded input









and output may cause serious problem in the presence of large noise or outliers since the

wrong predictive model due to noise may cause poor control. Hence, the selection of the

right model is as important as creating models and designing controllers.

Once the right local linear model is determined, the corresponding controller is

designed using (3.1). A schematic diagram of the proposed SOM based inverse control

system is shown in Figure 3-3 where the inverse control seeks to model the inverse of the

plant. A set of controllers appears in series with the plant. The command input, rk+, is

fed to the controller and provides also the desired response. Hence, when the error is

small the controller transfer function is the inverse of the plant.



11-- Plant P(z) No



. . ... ............................................
[ Embedding \- --- Embedding *






hN = gN ........



Figure 3-3. Proposed SOM-based inverse control scheme.

Generally, an adaptive controller that meets the specifications is slow to adapt.

However, our approach models all the operating regions and automatically divides the

operating regions by the number of PEs. So once the current operating region is

determined by the SOM, the corresponding controller is triggered so that the plant tracks

the desired signal. Moreover, even if the wrong PE is assigned in the winning PE due to

noise, a similar dynamic model can be activated since neighboring SOM PEs represent









neighboring regions in the dynamic space. Thus, the proposed control system can reach

the set point quickly, and even if the dynamic model is not the most appropriate, there is

an extra flexibility to match the set point with the least amount of error.

3.4 Multiple PID Control

In linear control theory, despite the development of more sophisticated control

strategies, Proportional-Integral-Derivative (PID) controllers have been extensively

studied by researchers and well understood by practitioners, since they are widely used in

practice, and their principle is well understood by engineers [95,92]. It gained its

popularity for its simplicity of having only three parameters. But owing to its simplicity it

has also paid the price of not having an efficient and practical way of determining

optimal gains.

The ideal continuous time PID controller is expressed in Laplace form as follows:

K
G(s)= Kp + K +Kds (3.2)


where KP is the proportional gain, K, the integral gain, and Kd the derivative gain.

Each of the terms works "independently" of the other1. The standard PID control

configuration is shown in Figure 3-4. The introduction of integral action facilitates the

achievement of equality between the measured value and the desired value, as a constant

error produces an increasing controller output. The derivative action indicates that

changes in the desired value may be anticipated, and thus an appropriate correction may

be added prior to the actual change. Thus, in simplified terms, the PID controller allows

contributions from present, past and future controller inputs.

1 This is not exactly true since the whole thing operates in the context of a closed-loop. However, at any
instant in time, this is true and makes working with the PID controller much easier than other controller
designs.









G(s)


R(s) E(s) U(S) Y(s)
I-- C--- -K=s- P(s)






Figure 3-4. PID controlled system.

Extensions of the PID control methodology to nonlinear systems, however, are not

trivial. Usually, global nonlinear models are, as always, linearized and the parameters of

the PID controller are scheduled according to the regime. All these difficulties, in fact

stem from the fact that the system model is constructed from a nonlinear dynamical

equation. This difficulty could be eliminated by a piece-wise linear approximation of the

nonlinear dynamics (as one does in gain scheduling). However, gain scheduling methods

are not flexible for model uncertainties and they cannot be scaled up to regimes that are

not described in the initial system identification stage by a linear model. It also has the

problem of either inefficient use of system identification data due to a large number of

arbitrarily selected operating (linearization) points or inaccurate modeling due to the

small number of linearization points [83].

The multiple PID control design method described in this section is rooted in the

principle of using local linear models to construct a globally nonlinear (piecewise linear)

system model that is determined completely from the input-output data collected from the

actual plant [12,30]. In addition to previously designed multiple inverse control schemes,

this section illustrates how the multiple models can be united with the well-known linear

PID controller design techniques to obtain a principled and simple nonlinear PID









controller design methodology. The proposed closed loop scheme is illustrated in Figure

3-5 where the SOM determines which linear PID controller contributes to the

instantaneous control input.


G- l(z) SOM -- Embedding --

SG (z ) P, ) kY k+ 1



rk+1



Figure 3-5. Overall schematic diagram of the nonlinear PID closed loop control
mechanism using multiple controllers.

Once system identification is complete, the design of a globally piece-wise linear

PID control system can be easily accomplished using standard techniques. The literature

has an abundance of PID design methodologies for linear SISO systems including direct

pole-placement techniques and optimal coefficient adjustment according to some criteria

[95,92]. Here pole-placement technique [4] is utilized to design a PID controller for each

linear SISO model and is illustrated briefly in the following.

Assume that the plant is modeled by a set of the SISO, second-order system given

by

Yk+ = f(YkYk -l,kik ,k)
= a ,,Yk + a2,,Yk 1 + b,k +b 2,uk 1

where Yk is the output of the plant, and uk is the input to the plant. The controller is

designed by starting with a general PID regulator shown in Figure 3-6 and determining









the coefficients of polynomials C, and D, so that the closed-loop system has desired

properties.


R(z+1) E(z) DU(z) ---Y(z +1)

S( (z) F, ( (z)z)




Figure 3-6. Block diagram of PID controller for a SISO plant model.

Each model's transfer function is given by

B, (z) b, z2 + b2~, z
(z)= -=2 (3.4)
A (z) z al,, -a2,,

It is assumed that the polynomials A, (z) and B, (z) do not have any common factors.

Note that in order to have a stable zero (zero inside the unit circle), b,, > b2, must be

true. The controller specifications are expressed in terms of a model that gives the desired

response to command signals. The general control equation is

C, (z)U(z) = D (z)(R(z) Y(z)) (3.5)

It is assumed that the C, is monic. To make sure that low-frequency disturbances give

small errors, C, is chosen as

C (z) = (z 1)L C (z) (3.6)

with a suitable selection of L. This puts a (z-1) term in the denominator of the

controller transfer function, guaranteeing integral control action.

The goal of the controller design is to map the values of a b,,,a2,Z,b,, and b2,, into

the controller coefficients, d,, and c, subject to the constraint of the model whose









characteristic polynomial is z2 + IlZ + 2 The transfer function of the closed-loop

system is 1/(l+zA(z)C,(z)/B (z)D (z)); therefore, its characteristic polynomial is

A (z)C, (z) + z 1B, (z)D, (z). The characteristic polynomial for the model is Am, with the

addition of a second-order "observer polynomial" term Ao. Equating characteristic

polynomials results in the design equation

A (z)(z-1)L C, (z) + z-'B, (z)D (z) = A, (z)Am(z) (3.7)

which can be written as

(z2 a,z -a)(z 1)(z +c,) + z-(b,,z2 +b ,,z)(d,,z2 + dz,,) = z2(z2 + z + A2)(3.8)

This equation is solved for the d,, and c,. Then the control law difference equation is

uk =(1 -)uk-1 +c,uk-2 + d,,ek + d2,,ek-1 +d3,ek-2 (3.9)

Given the local linear models as obtained through the use of a SOM and a PID

design technique, the overall closed loop nonlinear PID design reduces to determining the

coefficients of the individual local linear PID controllers using their respective linear

plant model transfer functions. One needs to determine a set of PID coefficients per linear

model. In the competitive SOM approach, the model output depends only on a single

linear model at a given time; therefore, the PID coefficients are set to those values

determined for the instantaneous winning model.














CHAPTER 4
MULTIPLE QUASI-SLIDING MODE CONTROL

Attempts to use traditional control methods, such as inverse control and PID

control, for nonlinear plants will inevitably encounter problems when faced with the

nonlinear nature of these systems. In order to overcome these difficulties in designing

controllers for nonlinear systems, a simplified control method that keeps the advantage of

the conventional approach was proposed in Chapter 3, i.e., the SOM was explored as a

modeling infrastructure, and the controllers were built based on SOM-based multiple

models.

However, these control methods (especially multiple inverse control scheme) may

show poor control performance when existing sensor noise or external disturbances due

to the fact that the perfect control is achieved if the plant and the controller is stable, if

the model is perfect, and if there is no disturbance. Wrong selection of the model caused

by noise can devastate the control mission since the controller is also determined by very

different model from the given task. For this reason, sliding mode control architecture,

which is a very well-known robust controller, is employed in this chapter in order to

obtain sturdiness against noise as well as uncertainties on the model.

4.1 Introduction to Variable Structure Systems

The control of nonlinear systems has been an important research topic and many

approaches have been proposed. While classical control techniques have produced many

highly reliable and effective control systems, great attention has been devoted to the

design of variable structure control systems (VSCS). Variable structure systems (VSS)









are a special class of nonlinear systems characterized by a discontinuous control action

which changes structure upon reaching a set of switching hyperplanes during the control

process to attain improved overall characteristics in the controlled system. During the

sliding mode the VSCS has invariance properties, yielding motion that is remarkably

good in rejecting certain disturbances and parameter variations [93,9,25,86].

Most of the VSCS proposed in the literature have been developed mainly based on

the state-space model with the assumption that all state variables are measurable or on the

input-output model for a linear system. But in some control problems, we are allowed to

access only the input and the output of the nonlinear plant [72]. In this case, an observer

could be used to estimate the unmeasurable state variables if the state equations are

known. Otherwise, this is not possible. Thus, it is the purpose of this chapter to provide a

new technique to design sliding mode control systems based on input-output models of

the considered discrete-time nonlinear system so that the amount of guesswork1 is

reduced, while attainable performance is increased.

Normally, the design of VSS consists of two parts: First, the sliding surface, which

is usually of lower order than the given process, must be constructed such that the system

performance during sliding mode satisfies the design objectives, in terms of stability,

performance index minimization, linearization of nonlinearities, order reduction, etc.

Second, the switched feedback control is designed such that it satisfies the reaching

condition and thus drives the state trajectory to the sliding surface in finite time and

maintains it there thereafter [93].



1 In most cases, we need to estimate the unknown parameters, unmodeled dynamics and bounded
disturbances. Also, it should be noted that the VSCS works best when the plant is completely known.









4.1.1 Sliding Hyperplane Design

Sliding surfaces can be either linear or nonlinear. The theory of designing linear

switching surfaces for linear dynamic system has been developed in great depth and

completeness. While the design of sliding surfaces for more general nonlinear systems

remains a largely open problem. For simplicity, we focus only on linear switching

surfaces. Moreover, for surface design, it is sufficient to consider only ideal systems, i.e.,

without uncertainties and disturbances. Consider a general system

= A(x)+B(x)u (4.1)

with a sliding surface

S = {x I S(x) = 0}

where A(x), B(x) are general nonlinear functions of x, and x e 91", u e 91'.

The equivalent control is found by recognizing that S(x) = 0 is a necessary

condition for the state trajectory to stay on the sliding surface S(x)= 0. Therefore,

setting S(x) = 0, i.e.,


S(x)=(L' )k = A(x) + -B(x)u =0 (4.2)


yields the equivalent control


Ueq B(x) A(x) (4.3)

where (OS/ 8x)B(x) is nonsingular. When in sliding mode, the dynamics of the system is

governed by


x= I-B(x) sB(x) s A(x) (4.4)
1 \^x ) x









For example, if the system (4.1) is linear and described by

S= Ax+Bu (4.5)

where A and B are properly dimensioned constant matrices. The switching surface can

be defined as

S(x) = CTx = 0 (4.6)

i.e., OS / Cx = C', where C = [c, c2, *, c, ] is an m x n matrix, and then we have

Ueq = (Cr'B) C'Ax (4.7)

and (4.4) becomes

= (I B(CT B) C' )Ax = (A BK)x (4.8)

(4.4) and (4.8) describe the behavior of the systems (4.1) and (4.5), respectively, which

are restricted to the switching surface if the initial condition x(to) satisfies S(x(to)) = 0.

For the linear case, the system dynamics is ensured by a suitable choice of the feedback

matrix K = (CTB) C'TA. In other words, the choice of the matrix C can be made

without prior knowledge of the form of the control u.

4.1.2 Sliding Mode Control Law Design

Once the sliding surfaces have been selected, attention must be turned to solving

the reachability problem. This involves the selection of a state feedback control function

u : 91" -> 9T" which can drive the state x towards the surface and thereafter maintains it

on the surface illustrating in Figure 4-1. In other words, the controlled system must

satisfy the reaching conditions.

The classic sufficient condition for sliding mode to appear is to satisfy the

condition s, s <0, i= 1,---,m and a similar condition proposed by Utkin [93], i.e.,









lim < 0 and lim s, > 0. These reaching laws result in a VSC where individual
S, _>0+ S, ->CT

switching surfaces and their intersection are all sliding surfaces. This reaching is global

but does not guarantee finite reaching time.


x











Reaching phase -- Sliding phase -- Reaching phase


Figure 4-1. Phase plane plot of a continuous-time second-order variable structure system.

Another commonly used reaching law is proposed by Gao & Hung [27]. The law

directly specifies the dynamics of the switching surface by the differential equation

S = -Q sgn(S)- Rg(S) (4.9)

where the gains Q and R are diagonal matrices with positive elements, and

sgn(S) = [sgn(s ),... sgn(s )] g(S) = [gl (s,),..., g (Sm )] where

sgn(s,) ={1 if s,(x) >0, 0 if s, (x) = 0, -1 if s,(x) <0} (4.10)

and the scalar functions g, satisfy the condition

s,g, (s,)> 0, when s, 0 (4.11)

Various choices of Q and R specify different rates for approaching S and yield

different structures in the reaching law.









For one way of designing controllers, recall that during sliding mode, one can

compute the equivalent control ueq according to (4.3) or (4.7). However, only using Ueq

cannot drive the state towards the sliding surface if the initial conditions of the system are

not on S. One popular design method is to augment the equivalent control with a

discontinuous or switched part, i.e.,

u= Uq + UN (4.12)

where ueq is a continuous control defined by (4.3), and uN is added to satisfy the

reaching condition which may have different forms. For a controller having the structure

of (4.12), we have





as
S(x) = x k A(x) + B(x)(ug + u)]

= [A(x) + B(x)ue q+ -B(x)uN

= -B(x)u,


for simplicity, assume (OS/8x)B(x) = I, then S(x) = UN. Some often used forms of UN

are relay type of control uN = -a sgn(S(x)), linear continuous feedback control

UN = -a(S(x)), and linear feedback control with switching gains UN = Yx where

S= [VI ] is an m x n matrix and

if s, (x)x > 0
Parameters and are chosen to ify the desired reaching condition.< 0

Parameters a and 8? are chosen to satisfy the desired reaching condition.









Another design method of controllers is to employ the reaching law approach

proposed by Gao & Hung [27] and can be directly obtained by computing S(x) along the

reaching mode trajectory, i.e.,

S(x) =-x= s[A(x) + B(x)u]= -Q sgn(S(x)) Rg(S(x)) (4.13)
ox ox

Hence, we have


u = B(x A(x) + Q sgn(S(x)) + Rg(S(x)) (4.14)


By this approach, the resulting sliding mode is not preassigned but follows the natural

trajectory on a first-reach-first-switch scheme. The switching takes place depending on

the location of the initial state.

4.2 Sliding Mode Control in Sampled-Data Systems

The VSS theory which was originally developed from a continuous time

perspective has been realized that directly applying the theory to discrete-time systems

will confront some unconquerable problems, such as the limited sampling frequency,

sample/hold effects and discretization errors [2]. Since the switching frequency in

sampled-data systems cannot exceed the sampling frequency, a discontinuous control

does not enable generation of motion in an arbitrary manifold in discrete-time systems

[3,9]. This leads to chattering along the designed sliding surface, or even instability in

case of a too large switching gain [2]. Figure 4-2 illustrates that in discrete-time systems,

the state moves around the sliding surface in a zigzag manner at the sampling frequency.

Much research has been done in this field. Among various concepts of discrete-time

sliding mode, Quasi-Sliding Mode (QSM) is reviewed in this section for the sliding mode

controller design aimed at sampled-data systems.









State trajectory

k
k+1
k+2


Sliding surface s = 0

Figure 4-2. Discrete-time system response with sliding mode control.

4.2.1 Quasi-Sliding Mode

Many researchers have either addressed the limitations when direct implementation

is done or have proposed designs that take the sampling process into account.

Milosavljevic [57] was among the first researchers to formally state that the sampling

process limits the existence of a true sliding mode. In light of this, the concept of QSM

has been suggested and the conditions for the existence of such mode have been

investigated. Consider a sampled-data SISO system with the predefined sliding surface

xk+1 =Axk + Buk (4.15)
Sk = CTXk = 0

The desired state trajectory of a discrete-time VSC system should have the

following features: Firstly, the trajectory moves monotonically towards the switching

manifold and crosses it in finite time starting from any initial point. Secondly, the

trajectory crosses the manifold in succession after it hits the manifold, resulting in a

zigzag motion about the switching surface. Lastly, the trajectory keeps on within a

specified band without increasing the size of each successive zigzagging step. Gao et al.

[28] defined a QSM as the motion of a discrete VSC system satisfying last two features.

In addition they named the specified band, {xI -A < s(x) < +A}, which contains the QSM

as the Quasi-Sliding Mode Band (QSMB). For single input systems, the main









approaches for the design of QSM control laws can be categorized into the following two

methods:

Discrete Lyapunov function based approach: Sarpturk et al. [81] noticed that unlike

the case in continuous-time SMC, the switching control in the discrete-time case should

be both upper and lower bounded in an open interval, in order to guarantee the

convergence of sliding mode. Recall that in continuous-time SMC, the control (4.12) is

composed of the equivalent control and a switching control. Converting this control to

discrete-time gives

uk = k,eq + k,N (4.16)

Hui & Zak [36] observed that if uk,N is a relay control with a constant amplitude,

the relay must be turned off in some neighborhood of the surface, in order to reach the

switching surface, otherwise, the trajectory will chatter around the surface with a chatter

amplitude at least as large as the amplitude of the relay output. The idea of sliding sector

[26,48] was used to solve this problem, i.e., to specify a region in a neighborhood along

the sliding mode, where linear control is used to keep the state inside the region after it

has reached the region. The switching control is applied only when system states are out

of the region. In this case, the derived switching surface is different from the sliding

surface. Based on a discrete Lyapunov function,

12
Vk-2k

the reaching law is given by

1
Sk(Sk+1 Sk ) < (Sk+1 Sk)2 for Sk 0 (4.17)
2


which ensures Vk+1 < Vk. Furuta [25] proposed a control law of the type









1k = uk,eq+FDXk (4.18)

where the equivalent control uk,eq is the solution of

ASk = Sk+1 Sk = 0 (4.19)

and therefore the equivalent control for the system (4.16) is

Uk,eq = (C 'B) 1 C(A- I)xk (4.20)

FD is a discontinuous control law which will be zero inside the sliding sector.

Reaching law based approach: Gao et al. [28] pointed out that the reaching law

(4.17) was incomplete for a satisfactory guarantee of a discrete-time sliding mode, since

it does not ensure that the trajectory moves monotonically towards the switching surface

and the trajectory stays on within a specified band. Thus, they presented an algorithm that

drives the system state to the vicinity of a switching hyperplane in the state space, rather

than to a sector of a different shape. They specified desired properties of the controlled

systems and proposed a reaching law based approach for designing the discrete-time

sliding mode control law. The equivalent form of the reaching law for discrete-time SMC

extended from the continuous-time reaching law (4.13), and for a SISO system is

sk+i -Sk =-qTsgn(sk)-rTsk, r > 0, q > 0, 1-rT > 0 (4.21)

where T > 0 is the sampling period. The state reaches the switching surface at a constant

rate -qT and the term -rT forces the state to approach the switching surfaces faster

when sk is large. The inequality for T guarantees that starting from any initial state, the

trajectory will move monotonically towards the switching surface and cross it in finite

time. Then the control law for discrete SMC is derived by comparing

sk+ sk = Cxk+ C'xk = C' Ax, + C' Buk -C'xk (4.22)









with the reaching law (4.21), which yields,

uk =(CB) [CT Axk -C k +qTsgn(sk)+rTsk] (4.23)

4.2.2 Quasi-Sliding Mode Control Using Multiple Models

Now we discuss the design of the control law for local linear models using the

QSM control framework proposed by Gao et al. [28], where the system states move in a

neighborhood around the sliding surface sk = 0. The central advantage of the sliding

mode control strategy is that it is an effective robust control strategy for incompletely

modeled or uncertain systems. Thus, the feature of the proposed control scheme is that

the robustness for disturbances can be obtained by the simple control logic based on the

linear model for each region. Another feature of the strategy is that it guarantees

convergence of the system output to a vicinity of the predetermined, fixed plane in finite

time, specified a priori by the designer.

Consider one of the local single input-output models f, of the plant f,

Yk+ = alk + a2yk 1 + + amYk-n+1 + bluk +b2uk-1 + + b,,uk-n+1 (4.24)

Equivalently, the input-output model of the plant in (4.24) can be written as the state-

space model2

xk+l = (k +Alk +A2Uk 1 + +AUk-n+1 (4.25)

where xk =[Yk-m+l"' Yk-1, Yk ] C 9 is the system state vector which is available for

measurement and cD and A1, -*, A, have the following forms:




2 In a formal state-space model, past values of the input should be included in the state vector using delay
operators. For simplicity, we include only the system output's past values in the state vector in this
notation.









0 1 0 0 *.. 0
0 0 1 0 *.. 0


0 0 0 0 ... 1
am am,1 am2 m-3 ... a


0 0 0

A1 = : 2 : ,A :
0 0 0



Also defining the tracking error vector as

ke+1 = k+ Xk+ (4.26)

where the desired signal vector is Fk+ = [dk-m+2, "..,dkdk+ ], the switching surface is

defined in the space of the tracking error vector given by

k = k (4.27)

where c =[c1,c2, ..,cm ]T. Then an equivalent control is designed to satisfy the ideal

quasi-sliding mode condition, sk+1 = sk = 0, by

ueq=(c TA) r(Fk+--1 Ak u)-TAzUk cTAuk.+} (4.28)

and the closed-loop system response of the ideal quasi-sliding mode substituting (4.28)

into with an equivalent control is given by

Xk+1 = -A,(cTA1) A J} k +A,(JTA1) rT k+1 (4.29)

The system (4.29) can be viewed as a linear system with the input Fk+~ and the output

k+I To get an insight into the tracking capability of the system, (4.29) can be

represented in terms of the tracking error e, = dk -Yk by









ml1 (4.-30)
ek+l -- ek -- ek-1 1ek-m+2 (4.30)
m m m

Note that by designing the switching surface such that the roots of polynomial

Atm 1 + (Cm_ /cM )A"' + -- + (c, / cm) are inside of the unit circle, the error vanishes and

thus the condition ensures asymptotic convergence to the desired output. An arbitrary

positive scalar cm also determines the time taken to reach the sliding surface and can be

adjusted to get a faster response.

The reaching law (4.21) always satisfies the reaching condition such that the

discrete VSC system designed using the reaching law approach is always stable with a

stable ideal quasi-sliding mode [28]. Then the control law is derived by comparing

sk+1 sk = Ck+l ck

with the reaching law (4.21), which yields,

uk = (T A1) [jTrk+l _T)k _T A2Uk .c A Uk-n+l (4.31)
+(rT- 1)sk + qTsgn(sk)]

Salient feature of the multiple quasi-sliding mode controller is that one can obtain

faster convergence to get the desired response due to multiple control scheme and one

can employ VSCS to control unknown nonlinear plants while gaining indemnity against

noise and parameter variations.

4.3 Analysis of Multiple Quasi-Sliding Mode Control with an Imperfect Sensor

For the SOM-based system identification, one needs to quantify the effect of the

modeling error that will occur due to the quantization of state space induced by the SOM,

and also by the wrong selection of the winning model. Consider model (4.24) in the

presence of modeling error and measurement noise. The predicted output becomes:









k+l = al (Yk + k ) + a2 (Yk-1 -+ -1 ) + m+ a (Yk-.m+l- + k+l )
Il 1 1 (4.31)
+ buk +b2 k1+---+ buk-n+l

where a wrong local model (^i, a2,, m* 1, ,, j is triggered by the SOM due to

the noisy output measurement Yk + k. Then, when sk =0, the overall tracking error

response with an equivalent control is given by

Cmr1 Cmr2 C1
ek+l k 2 k-k 1 ek-m+2
m m m
+ (a1 )(Yk + k ) +m + (am am )(Ykm+ + km+1) (4.32)
S(b2 -2 )k 1 -+ + (bn ,)uk -n+l

For simplicity, consider the error dynamics (4.32) when m = 2. Defining model

parameter error Ak = [a al, a2 a2,b -b ] and noise = [k, Ck k2 ]T

ek+1 =cek +Ak(k +fk) (4.33)

where zk =[Yk,Yk-l Uk-1]T, C =1 C2, and c is chosen as c <1. We assume that

S2 2 2
[]= 7 and Ak
between the reference vector of the correct PE and that of the neighboring PE selected by

the perturbed output measurements3. Also, it is assumed that the noise sk is zero-mean

and white. We have the following recursive formula for the tracking mean squared error.

E[e] =E[c2e 2Cek (k + ik ) k k )(k -k T (4.34)n
2E[e +AzkzAk +ATE[iik <]Ak

When we take the norm on each side in (4.34), the norm of the tracking error power is

represented by


3 The first assumption states that measurement noise has finite power. The second assumption means model
parameter error is bounded by the distance in the state space.









E[e2 1]< cE[e2]+ Ak 2k +2 Ak (4.35)

where the Cauchy inequality is used on the second term on the right hand side. As

k -> oo, using the earlier assumptions on noise and model error bounds, the following

bound on the steady-state tracking error power is obtained:

^ 2
E[e ]< Ik k [k 2 +- (4.36)
(1- cU2)

Note that the difference between the true model (winning PE) and the neighboring

model (wrong PE) assigned by noisy input, c 1-k -k is typically small, since

neighboring SOM PEs represent neighboring regions in the dynamic space. Also, it

should be noted that the error can still be very large if we choose c as close as 1. In

contrast, by choosing c as small as possible, the closed-loop system may have very fast

transient response, possibly too large unexpected overshoot. Thus we should be careful

for determining c so as not to have large error. This problem will be discussed later in

simulation results. If c is set to small enough it then follows that the error by choosing

appropriate design parameters mentioned above will be bounded for a given modeling

uncertainty and measurement noise bounded by y. Moreover, this shows that the

switching scheme does not create an issue to be considered in order to guarantee BIBO

stability of the overall system.














CHAPTER 5
CASE STUDIES

To examine the effectiveness of the proposed multiple controller design

methodologies, the chaotic systems with the input term, discrete-time systems, and flight

vehicles have been considered assuming the following:

Assumption 1: The only state available for measurements is yk = x1l.

Assumption 2: The nonlinear function f is completely unknown.

By assuming that the function f is unknown, we confront a worst case (least prior

knowledge) control design. If an estimate f of f is available, it can be included in the

control design. As can be expected, the better the estimate f the better the performance

of the resulting controller. Our objective is to design multiple controllers for unknown

nonlinear plants that guarantees global stability and forces the output, Yk, to

asymptotically track the desired signal, i.e., Yk rk 0, as k oc without any a priori

knowledge of the plant.

5.1 Controlled Chaotic Systems

Chaotic systems exhibit irregular, complex, and unpredictable behavior that exists

in many industrial systems. Recently considerable effort has been in the focus of attention

in the nonlinear dynamics literature since removing chaos can improve system

performance, avoid fatigue failure of the system, and lead to a predictable system

behavior. In the literature, several design techniques have been applied for the control

and synchronization of a variety of chaotic systems. A generalized synchronization of









chaos via linear transformation [106], adaptive control and synchronization of chaotic

systems using Lyapunov theory [103,107], nonadaptive and adaptive control systems

based on backstepping design techniques [31,32,99,100], and an adaptive variable

structure control system for the tracking of periodic orbits [108] have been considered, to

name a few. The contribution of this work lies in the design of multiple control system

for the control of chaos based on the theory of the inverse control, PID control, and

sliding mode control [14] assuming that the chaotic system is unknown. Given an

unknown chaotic system, the goal is to force it to set points or a stable trajectory.

5.1.1 The Lorenz System

The Lorenz model is used for fluid conviction that describes some feature of the

atmospheric dynamic [73]. The controlled model is given by

1q = (x x)
x2 = -x x-x3 x 1 + u (5.1)



where x,, x2, and x3 represent measurements of fluid velocity and horizontal and vertical

temperature variations, respectively. The Lorenz system can exhibit quite complex

dynamics depending on the parameter values. For 0
equilibrium point. For 1 < a < a* (a, ) := a(a + f + 3)/(o -/ 1), the system has two

stable equilibrium points + (a- 1),+ (a- 1),(a-1)) and one unstable equilibrium

point at the origin. For a* (o,/ ) < a, all three equilibrium points become unstable and

the system trajectory have chaotic behavior [109]. As in the commonly studied case, we

select = 10, = 8/3 which leads to a *(o,/7)= 24.74. Thus we will consider the

system with o- = 10, / = 8/3, and a = 28, which produces the well-known butterfly











chaotic dynamics without any control, i.e., u = 0, as shown in Figure 5-1. Here, the


control force, which is uniformly distributed in u < 100, is added into the second


equation of (5.1).


20



1 "' 3 5 10 15 20 25 30 35 40 45 50


'i 0 ~r A / ~ i AA X .
0 ----------------------------
6 5 I0 15 20 25 30 35 4D 45 50


520
400"_
2 2 4 00

x2 -40 0 O 5 10 15 20 25 30 35 40 45 50
x3 Time (sec)


Figure 5-1. The uncontrolled Lorenz system: phase-space trajectory and time-series.

After solving the set (5.1) with forth-order Runge-Kutta method with integral step

0.05, 7000 samples were generated for analysis. In order to construct a set of model-

based controller, first, we design a model as


k+ = f(Yk,Yk-1 ,Uk,k-, ,Uk- 2) (5.2)

based on the Lipschitz index [35] as shown in Table 5.1.

Table 5-1. Lipschitz index of the controlled Lorenz system for determining an
embedding dimension.


Number of inputs

1 2 3 4 5
1 1090.9 49.7 5.7 3.3 2.1
Number 2 10.5 2.9 2.4 2.1 1.6
of 3 5.1 2.7 2.0 1.5 1.3
outputs 4 3.2 2.3 1.9 1.4 1.2
5 3.0 1.8 1.4 1.4 1.2


-10^







60




0 OS
0 00
0 045
0 04 40
0 035
30
SD03-
0 025 20
002
0 01510
001 01
05 10 15 0 500 1000 1500 2000 2500 3000 3500 4000
Number of PEs of one-side ofa map Epochs


Figure 5-2. Generalization error v.s. Number of PEs (left) Learning curve (right).


The SOM was trained with the output history vector V/y,k I = Yk- 1] over


L = 6000 samples with the time decaying parameters, 7k = 0.1/(1+0.003k) and


Ck = (,[N / 2)/(1+ 0.003k) in (2.14), and then local linear models were built from V/y,k


as well as V',,k = U k-1 ,Uk- 2] for each PE. A newly generated sequence of M = 1000


samples was applied to multiple models for performance test1. Then the dimension of the


square SOM was determined as 8x8 based on the generalization error as shown in Figure


5-2 since the error does not decrease much after N = 8. Also, with N = 8 x 8, the


learning curve2 for 4000 epochs is as shown in Figure 5-2. This curve reflects the overall


closeness of the winning PE to the input samples during the training process and it


becomes approximately constant at the end of training.







Identification performance was evaluated by NRMSE (Normalized root mean squared error):

= 1/ max(r) 1/M IZ (rk+1 k+ )2

2 The learning curve is defined as the RMS (Root mean squared) distortion between the input and the

winning PE: = 1/L z |1yk W,k







61


The identification result is shown in Figure 5-3 where the dashed line is the output

of the controlled Lorenz system, and the solid line is the output from the multiple models.

As we can see, the multiple models are a very good approximation of the controlled

Lorenz system. In addition, plant modeling performance with 64 multiple models was

compared with a single linear model, ARX, with the same number of inputs used in local

modeling.


i-20---------

S 5 10 15 20 25 30 35 40 45 50

UJ
-10







0




0 5 10 15 20 25 30 35 40 45 50
Time (sec)


Figure 5-3. Identification of the controlled Lorenz system by multiple models.

It was also compared with those by means of a conventional TDNN, which was

trained by the backpropagation algorithm with the constant learning rate of 0.001 for

3000 iterations on the same number of inputs and outputs as in local linear modeling,

adopted as a global nonlinear model and listed in Table 5.2. The best result with the

proposed method was a NRMSE of 0.0131 while with the TDNN3 obtained a NRMSE of



3 The number of PEs in the hidden layer of the TDNN is chosen as 10 by 20 Monte-Carlo simulations
varying the size of the hidden layer.









0.0205. On the other hand, a single linear ARX model produced much higher NRMSE,

0.0486, than others did. This result shows that the proposed multiple linear modeling

scheme outperforms both the linear and the nonlinear global modeling paradigm.

Table 5-2. Comparison of modeling performance for the controlled Lorenz system.

Methodology NRMSE

ARX (1) 4.8e-2
Multiple ARX (64) 1.3e-2
TDNN (5:10:1) 2.1e-2

Next, we tested the proposed control scheme using multiple models. First, we let

the system converge to the origin, which is one of the equilibria of the Lorenz system,

starting from the initial state [x, x, x3] = [10,10,10]' by the Multiple Inverse Controllers

(MIC); the controller is activated at 5 sec and the results are shown in Figure 5-4(a)

where the closed loop system response is seen to converge to the origin fairly well. This

is relatively easy to control since the affine system (5.1) of zero dynamics is

asymptotically stable at the equilibrium point. Also the figure shows the behavior of the

control input. Second, we forced the controlled Lorenz system to a set point, Yd = 8

which is not one of the equilibria. This is a more complicated task than steering the state

of the system to the origin. The regulation results with the proposed MIC are shown in

Figure 5-4(b) where we observe that the states x, and x2 asymptotically regulate to

x, = 8 and x = 8, respectively, and the state x3 remains bounded. They illustrate that the

conditions x2 = x1 and x3 = (x,)2 /1/ described in [108] are satisfied even in the case


under consideration.








63



50 100

40 -, 2 00

30 1 "

20 40



040
-100
-60

20 -80

2 3 4 5 7 910 0 1 2 3 4 5 7 9


Time (sec)

(a)

50 100
40 2 0
30
10 I 0 f' 40
'* t I 2' 0




40
-10
2o-20

20 -80
-30 -100


Time (sec)


2 3 4 5 B 7 8 9 10 1 2 3 4 5 6 7 B 9 10
Time (sec) Time (sec)

(b)


Figure 5-4. Tracking a fixed point reference signal by MIC: (a) yd = 0 (b) yd = 8 .


Afterward, the MIC was investigated as increasing the number of controller and


compared with a single linear inverse controller built by a ARX model as well as a


nonlinear TDNN controller, which is a global controller trained through the TDNN


model, for the same task. The optimal number of PEs in the hidden layer of the TDNN


controller was selected as 40. Figure 5-5 illustrates how the number of controllers effects


control performance. As expected, a single inverse controller based on a ARX model


showed the worst performance with regard to settling time as well as steady-state error


even though it reached the set-point the first time. As seen in the figure, the faster the


rising time, the larger the overshoot, which is an unwanted factor in most cases, is


E










generated. A smaller overshoot, faster settling time, and less steady-state errors were

obtained by using a larger number of inverse controllers, such as the 16-MIC, which

showed a very similar performance to the TDNN controller. Moreover, the response

using the 64-MIC demonstrated very small steady-state errors. The 144-MIC, however,

reached the set-point much slower than the others. It also took much longer to settle to

the desired point. Too many divisions of state space may cause poor control performance

(frequent switching among controllers) so that the controller may not be capable of

following fast-changing trajectory.


13
*** . .... TDNN
12 ---- Set-point -
MIC (1)
11 MIC (16)
MIC (64)
10 \-- MIC (144)




77 5.8 5




4





Figure 5-5. Comparison of control performance varying the number of inverse controllers
based on multiple models.

Finally, the proposed multiple control schemes, especially MIC, were compared

with a linear controller and a nonlinear controller. The PID controller coefficients were

determined to bring the poles of the closed-loop response from the plant output to the

desired output to 0, 0, 0.55+/0.31, and 0.55-/0.31. The QSM controller was designed by











choosing the switching surface with c = [1,-3]T and qT = 0.8, rT = 0.001. Depending on


the selection of the coefficients for PID and QSM controllers, one obtains the

performance desired. This issue will be discussed in later sections. The step performance

of the designed multiple model based control scheme and the TDNN controller in closed-

loop operation with the controlled Lorenz system is provided for the two case studies,


Yd = 0 and Yd = 8, in Figure 5-6.


TDNNC TDNNC
IC-ARX IC-ARX
MIC 12- MIC
-- MPIDC MPIDC
MQSMC -- MQSMC
10

8-






48 5 52 54 56 58 45 5 55 6
Time (sec) Time (sec)


Figure 5-6. Comparison of tracking performance by multiple model based controllers
(MIC, MPIDC, MQSMC) and global inverse controllers (IC-ARX,
TDNNC).

We can observe from these results that the overshoots of the Inverse Controller based on

the ARX model (IC-ARX) are much larger than others, which also results in a much

longer settling time. Moreover, even if IC-ARX and TDNNC demonstrated very short

reaching time, they exhibited relatively larger steady-state errors than the others. On the

contrary, the multiple inverse controller showed a much faster reaching time, which is

close to that by IC-ARX and TDNNC, with small steady-state errors when compared

with MPIDC and MQSMC. These comparisons of IC-ARX, TDNNC, and MIC seem

reasonable since they all have an inverse control framework. From these results, it can be

easily inferred that the proposed control strategy guarantees the convergence of the






66


system gaining a fast response to the desired set-point, even though the plant with the

highly nonlinear characteristics is not known a priori to the controller.

5.1.2 The Duffing Oscillator

Another system considered corresponds with the control of a Duffing oscillator,

which displays chaotic behavior, described by


S= pX2 3 (5.3)
x2 = 1 -pIx2 p2 p3I + P4 COS at

where 0 is a constant frequency parameter, p p2 ,P3 and p4 are constant parameters

[10,32,34,35].






I 2,_ I10 20 30 40 50 60 70 80 90 100





-2 6 1 0 5 0' 5 1 1 5 2 0 10 20 30 40 60 60 70 D0 90 100
x1 Time (sec)

Figure 5-7. The uncontrolled Duffing oscillator: phase-space trajectory and time-series.

We assume that the controlled Duffing oscillator is originally (u = 0) in the chaotic state,

shown in Figure 5-7, with parameters o = 1.8, [p, p2,p3,p4] = [0.4, -1.1, 1.0, 1.8]'

[32]. In the simulations, the Duffing system was considered as unknown, which only

generated time-series data excited by uniformly distributed control input, u < 5, via the

fourth-order Runge-Kutta scheme with a fixed time step of 0.2. The embedding

dimension for model construction was chosen as d, = 1 and d, = 1 based on the Lipshitz


index in Figure 5-8. Thus the model is designed as k+ = 1 f(Yk,Y ,Uk,Uk ) Another







67



parameter to be selected is the size of the SOM which is trained with f = [YkYk 1] T


We chose the number of PEs in the SOM as 8x8 by the performance varying the size of

the map shown in Figure 5-8.


Number of outpu s


4 Number of inputs


0036

O03

0 025

002

001

n n


5 5 2 4 6B 10 12
Number of PEs of one-side of a map


Figure 5-8. Lipschitz index (left) for the determination of optimal number of inputs and
outputs and Generalization error v.s. Number of PEs (right).

Table 5-3. Comparison of modeling performance for the controlled Duffing oscillator.


Methodology NRMSE

ARX (1) 3.4e-2

Multiple ARX (64) 1.3e-2

TDNN (4:14:1) 1.le-2


The identification result with 8x8 square map and its performance comparison with

one of global model, TDNN4, is presented in Figure 5-9. In addition, another comparison


with a single linear model is listed Table 5-3. We observe that the proposed multiple

modeling strategy is a little worse than a TDNN, even if it demonstrated much better

modeling performance than a single ARX model. The proposed modeling method,






4 A TDNN with 14 PEs in the hidden layer demonstrated best performance in modeling the Duffing
oscillator.










however, has a much simpler structure for modeling chaotic systems than a global

modeling paradigm.





4- 4-




A 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100


Time (sec) Time (sec)

Figure 5-9.Identification of the controlled Duffing oscillator by TDNN (left) and
multiple-models (right).

Next, we performed simulations for 3 different control tasks.

1) Control of a Duffing oscillator: For this task, we target the oscillator to follow an

arbitrary trajectory generated by a random control input bounded by 5.

2) Synchronization of two Duffing oscillators: This is to show the proposed control

scheme is able to synchronize oscillators effectively in spite of model mismatch; the

parameters for the slave oscillator are p, = 0.4, p4 = 1.8, and o = 1.8, whereas the

parameters for the master oscillator are p1 = 0.41, p4 = 2, and o = 1.9.

3) Synchronization of two strictly different second order oscillators: In this case,

the slave oscillator is a Duffing one. For this oscillator, the parameters were taken as

p = 0.4, p4 =1.8, and o = 1.8. The master oscillator is a van der Pol one which is


described by x, = x, 2 = -0.1(x,2 1)x -xi' +1.75 cos* (0.667t)

For these missions, the controllers were designed as follows: The optimal number

of PEs in the hidden layer of the TDNNC was selected as 30 by 20 Monte-Carlo









simulations. The MQSMC were built by choosing the sliding hyperplane with

j = [1,-15]' and qT = 0.8, rT = 0.001. The MPIDC were designed for placing the poles

of the closed-loop response at 0.25 + i0.25 which demonstrated the fastest convergence

to the desired trajectory in Figure 5-10.


Fime (sec) Time (sec)
(a) (b)


Time (sec) Time (se-)
(c) (d)

Figure 5-10. Performance comparison on trajectory tracking by TDNNC, PID-ARX,
and MPIDC when the poles of the closed-loop response are place at (a) 0.9
(b) 0.5 0.5i (c) 0.25 0.25i (d) 0.05 0.05i.

From these figures, it should be noted that the MPIDC produces shorter settling time than

a single PIDC does regardless where the poles are placed. In addition, as the poles are

getting closer to the unit circle or to the origin the convergence time is getting longer.

Hence, the poles of the error dynamics should be chosen carefully.









We compared the performance of the controllers using 50sec-long oscillatory

trajectory regarding settling time and NRMS of steady-state error (NRMS-SSE) in Table

5-4: the settling time was selected when the tracking error was bounded in +0.1 and the

NRMS-SSE was calculated using the tracking error from 30sec to 50sec. As seen in the

table, the proposed control strategies outperformed a global controller, which is generally

utilized for unknown system control, in both fast response and accuracy. Especially,

while the TDNNC demonstrated some difficulty in following the desired path generated

from a different dynamics, the van der Pol oscillator, the multiple controllers

accomplished the mission relatively well.

Table 5-4. Comparison of tracking performance for 3 different control task : Settling
time and NRMS-SSE.

Methodology Task 1 Task 2 Task 3
6.0sec 3.6sec 10.6sec
2.8e-2 3.2e-2 6.8e-2
5.0sec 3.6sec 7.8sec
2.3e-2 2.2e-2 2.7e-2
3.6sec 1.6sec 6.2sec
1.9e-2 1.9e-2 1.8e-2
4.2sec 2.4sec 5.2sec
MQSMC
2.7e-2 2.6e-2 3.2e-2

Moreover, the MPIDC among the proposed multiple control methodologies showed a

significant reduction of the transient time as well as accuracy improvement in most

missions. Figure 5-11 shows the synchronization of the master and slave oscillator

signals by the TDNNC and MPIDC where the controller is activated at t = 8 sec. It is

seen from the results again that the transient time is shortened without resulting in

overshoots using the MPIDC.




















-1


0 5 10 15 20 25









5 10 15 20 25
Time (sec)








-11

2
-3
0 5 10 15 20 25














0-3
0 5 10 15 20 25























a 5 10 15 20 25
Time (sec)












60 10 15 20 25









6 5 10 15 20 25
Time (sic)


ime (sec)


2

-3
0 5 10 15 20 25









0 5 10 15 20 25
Time (sec)



3


2



-3
0 5 10 15 20 25


10 15
Time (sec)


Control performance by TDNNC (left) and MPIDC (right): The dotted

line is the desired trajectory and the solid line is the output of the

oscillator. (a) Tracking an oscillatory reference signal (b) Synchronization

of two Duffing oscillators (c) Synchronization of a Duffing oscillator and

a van der Pol oscillator.


Figure 5-11.










5.2 Nonlinear Discrete-Time Systems

We have seen how effectively the PID and the inverse controller can be

implemented using multiple models. In this section, nonlinear discrete time systems in a

noisy environment are considered focusing more on multiple model-based control with

sliding mode.

5.2.1 A First-order Plant

Consider the following nonlinear discrete-time plant [5]

,k+l "2,k

x2,k+1 = l+ )2 +x2+uk (5.4)
16 1+ (x2,k 2

Yk = X2,k + Zk

where uk is the input and zk is an external disturbance. In (5.4), we considered a SISO


model, assuming that only the output Yk is available for measurement. The output time-


series was created by exciting an input signal that is uniformly distributed in u < 0.5.

x10









4 I
2 2 1
Number of o puts 1 1 Nu br of nputs 2 4 6 8 10 12 14 16

(a) (b)

Figure 5-12. Parameter selection to design multiple models: (a) Lipschitz index for
determining the embedding dimension (b) Identification performance v.s.
network dimension on independently generated test data for choosing the
size of a map.







73


The model was assumed to be a second order in input and output based on Figure

5-12(a). After quantization of the embedded output space V/,, a set of models was built


with the input-output data samples for each PE. For testing, 400 independently generated

data samples were used changing the size of the map. The best size of the map was

determined as 8x8 since the performance did not improve much after 64 PEs (see Figure

5-12(b)). Thus plant identification with 64 multiple models (8x8) was tested in the

absence of sensor noise as well as in the presence of sensor noise with the plant input

signal being uniformly distributed. The result of system identification in the absence of

sensor noise by multiple models is shown in Figure 5-13(a). As we can see, the models

provide a very good approximation of the plant visually based on the error signals.


S2---------------- 10',---------


E

2~ A
0 la 10 C 200 250 30 360 400



0 1 ~ --


TUN


S 560 100 160 200 250 300 360 400 0 6 10 15 20 25 30 35 40
Iterations SNR
(a) (b)

Figure 5-13. Modeling performance using 64 multiple models for a nonlinear first order
plant: (a) System identification of the nonlinear plant in the absence of
disturbance by the proposed multiple models. (b) Comparison of robustness
against noise between TDNN and multiple models.

Also, the proposed multiple modeling scheme was compared with a conventional

TDNN that was trained through the backpropagation algorithm with 5000 samples and a

constant learning rate, 0.005. The modeling result with 64 multiple models was a NRMSE










of around 6.5e-4 while with the TDNN5 one obtained a NRMSE of about 6.3e-3. The

performance when the testing data are perturbed by noise is shown in Figure 5-13(b)

where we observe that the SOM is more likely to select the wrong model as the noise

level is increased. Even though the multiple models presented more accuracy in

identification than a global model in the absence of noise, it should be pointed out that

the multiple modeling strategy does not have more noise-immunity than global models at

certain noise level. However, the proposed method is more robust than a global model up

to certain noise level.


-8- T6 -8- qT-05
-A- 7Tv-6 -E- qTf01
12 --e- T04 1 7-0
1 2 4 T 8








2 4 6 8 10 12 14 16 1 20 0 2 4 6 1 4 16 1 20
Iterations Iteratons

Figure 5-14. Responses for parameter selection to design QSMC by varying (a) rT and
(b) qT.

Based on the 64 multiple models and the TDNN model identified, we designed

multiple controllers and a TDNN controller, respectively. The TDNN controller was

trained by back-propagating an error through the TDNN model taught by 20 hidden PEs.

The number of hidden PEs in the controller was chosen as 40. The MPIDC was designed

for placing the poles of the closed-loop response at 0.1 i0.3. The MQSMC was built by


choosing the sliding hyperplane with c = [1,-2]T and qT = 0.8, rT = 0.01 comparing the


5 The number of PEs in the hidden layer of the TDNN is chosen as 20 by 20 Monte-Carlo simulations
varying the size of the hidden layer.











parameters as in Figure 5-14. In the figures, we observe that the larger the value of rT is


the faster the plant reaches to the reference signal. In addition, the smaller the value of


qT is the smaller the sliding mode band is.


First, the performance of the controllers was tested on square-wave [1,-1.5, 1, 0]


tracking in the absence of sensor noise and the results are presented in Figure 5-15 where


we can see that all controllers showed very good performance on tracking the reference


signal. Specifically, the MPIDC is the most accurate controller in spite of very slow


convergence, and the MIC is the fastest one even though it shows a little steady-state


error. In contrast, the MQSMC demonstrated the worst performance among the 4


controllers regarding rising time and steady-state error, but it shows its superiority in the


presence of noise later.


151 04
TDkNNC
1_ MIC
MP 1 02
MQSMC
05


0 98
D5 .' TNN


S94- M
MOSMC
0 92
0 10 20 3 40 50 60 70 8 90 100 34 36 38 40 42 44 46 48 50 52
Iterations Iterations


Figure 5-15. Comparison of tracking performance using a global controller and
multiple model based controllers in the absence of sensor noise. The figure
(right) is an enlargement of the figure (left) between 34 and 52 iterations.

Figure 5-16 shows the plant responses of the closed loop control system using the


MQSMC. The trajectories are seen to converge to the desired values of [-1.5 -1.0 -0.5 0.5


1.0 1.5]. The figure also shows control input, the sliding surface, and the winner activities


switched automatically by the SOM. It can be easily seen that the proposed MQSMC







76


scheme guarantees the convergence of the system to the quasi-sliding-mode band around


the sliding hyperplane ek = 0.


20 40 60 IIso 1
ltera


20 140 160 180 2


S2 20
-3_.!
10 -

0 20 40 60 O0 100 120 140 160 10 200 0 20 40 6O 80 100 120 140 160 180 200
Iteration Iterations


Figure 5-16. Performance of square-wave tracking in the absence of noise by the
MQSMC.

Next, the robustness of the proposed control scheme was compared with that of a


global controller using TDNN. The standard deviation of the error between the plant


output and the desired output versus the standard deviation of the noise is shown in


Figure 5-17. It is evident that the MQSMC performs best in terms of insensitivity to


disturbances. The MIC structure showed the best performance only in the noise-free


environment. It should be noted that the MIC and the MPIDC began to become less


robust than the TDNNC at the point where the standard deviation of the noise is over


0.04. From this examination we can conclude that the MIC and the MPIDC can be robust









against noise up to certain level. However, wrong selection of the winner, as the amount

of noise is increased, can be devastating for the controller that is designed based on the

predicted model.


2 TDNNC E MIC B MPIDC DMQSMC

b 0.18-
y 0.16 -
0.14-
0.12-
0.1
0.08
0.06-

1 0.024 H

0.000 0.003 0.011 0.023 0.046 0.070 0.093 0.115
Standard deviation of noise


Figure 5-17. Comparison of performance against noise among TDNNC, MIC, MPIDC
and MQSMC.

Furthermore we tested the closed loop system for tracking a sinusoidal and an

arbitrary desired output. Once again, the multiple controller networks perfectly track the

desired command except for a transient time of a few time steps, as shown in Figure 5-18,

even if the measurement is corrupted by zero-mean random noise with 20dB of SNR.

Overall, we conclude that the proposed MQSMC approach is the most robust

design technique among the four methods considered. This is evident from Figure 5-17,

where we observe that on average the tracking error of the MQSMC increases at a lower

rate than that of the MIC, the MPIDC, and the TDNNC.

















0 5 0 -
0 0
0 0
05 -05




15i--------------------- 15----------------------
L 20 40 El 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200
Iteration Iteration




05 09722 0.3578 0.1295 0.3103
0 0


4228-1 663 0.03259
-1 30 -2 U 5
-22

25 260 40 6O 80 100 120 140 180 180 200 0 20 40 60 B0 100 120 140 160 180 200
Iteration Iteration

(a) (b)

Figure 5-18. Sinusoidal and arbitrary signal tracking by the MQSMC:(a) in the absence
of sensor noise (b) in the presence of sensor noise, SNR = 20dB.

5.2.2This model is obtained through identification of a laboratory-scale liquid-leveLiquid-level Plant


A liquid-level system described by the following second-order equation has been

considered:


Yk+1 = 0.9722yk +0.3578uk -0.1295uk-1 -0.3103ykUk
-0.04228 k- +0.1663yk-1k-1 0.03259yk yk- 1

--0.03513y uk-1 +0.3084ykyk-lUk-I

+ 0.1087yk-lUkUk-1


This model is obtained through identification of a laboratory-scale liquid-level system


[80,1]. In [80], this model has been used to illustrate theoretical developments for direct


adaptive control. 10,000 data samples were generated for analysis using the control










effort, 'i <1, which has uniform distribution. Given the prior information concerning


the order of the plant by Lipschitz index in Table 5-5, a second order input-output model

described by the following equation was chosen to identify the plant:


k+l =f(Yk Yk-1 kk k-1)


(5.6)


Table 5-5. Lipschitz index of a laboratory-scale liquid-level plant
embedding dimension.


Number
of
outputs


for determining an


Number of inputs


262.34
13.49
1.46
1.19
1.19


10.86
1.32
1.16
1.15
1.15


1.80
1.18
1.13
1.10
1.09


1.64
1.14
1.13
1.10
1.09


1.46
1.13
1.12
1.09
1.09


liii..


4 6 8 10
Number of PEs of one-side of a map


12 14


01

1-0 104


50 100 150 200 250
Iterations


Figure 5-19.


Modeling performance using multiple models for a liquid-level plant: (a)
Identification performance v.s. network dimension for choosing the size of a
map of the liquid-level plant (b) Identification of a liquid-level plant using
12x12 multiple models.


The embedded output vector, Vy,k', was used for SOM training over 6,000 epochs,


and then not only V'y,k but the embedded input vector, V' ,k, were exploited to create


local linear models. In order to test the performance of the proposed modeling scheme,


0 025

002

0015T

001

0 005


00 350 400









we applied 400 newly generated data samples to multiple models. Figure 5-19 depicts

identification performance depending on the size of the network and how well the

multiple models approximate the liquid-level plant using 144 models.

In addition, we compared the identification performance using multiple models

with that by a TDNN. The number of inputs and outputs to the network were the same as

in local linear modeling. A single hidden layer of 30 PEs was large enough for good

identification performance by 20 Monte-Carlo simulations varying the size of the hidden

layer. As seen in Table 5-6, the proposed multiple models slightly outperformed a

nonlinear model in a liquid-level plant identification. Also the table clearly shows the

benefit of using multiple models when comparing with the performance by a single ARX

model.

Table 5-6. Comparison of modeling performance for the liquid-level plant.

Methodology NRMSE

ARX (1) 29.2e-3
Multiple Models (12x 12) 3.5e-3
TDNN (4:30:1) 4.3e-3

A typical open-loop input-output characteristic of the plant is shown in Figure 5-20.

The large variations in the steady-state gain and time constant with the operating point is

also clearly visible from this figure. The performance of the proposed multiple model

based controllers were compared to that of a nonlinear TDNN controller designed

through the previously identified TDNN model. The controllers for this plant were

designed as follows: First, the optimal number of PEs in the hidden layer of the TDNN

controller was chosen as 50. Second, the PID controller was designed in order for the

closed loop response poles to be located in 0.7702 and 0.1298. Third, the sliding








81



hyperplane of the QSM controller was set to j = [1, -1.66]' based on Figure 5-21 which


illustrates the effect of the parameter selection for the sliding surface design under


different noise levels.



0.6
-- Output
0.5 ---- Input

0.4 -

0.3

2- 0.2

0.1
0 - - - - - - - -



-0.1
I 0 \



-0.2 --

-0.3 -

0 100 200 300 400 500 600 700 800 900
Iterations


Figure 5-20. Typical input-output characteristic of the second-order liquid-level plant.



0045
SNR=40dB
004 -0 SNR=35dB
-B- SNR=30dB
0035 -*- SNR=25dB
-e- SNR=20dB
0 03 -

u 0 025 -

i 0.02

0.015



0 00


01 0.2 03 04 05 0.6 07 0 09 1
Icl/c21


Figure 5-21. Square-wave tracking performance of the liquid-level plant varying the
sliding surface and the noise level by the MQSMC.










In most cases, as we placed the poles (-c, / c2) closer to the origin inside the unit circle,

the controller showed better tracking performance. For instance, from the plot, we can

say that the pole should be chosen as between 0.5 and 0.6 to have the robustness against

noise whose level is 25dB of SNR since the error changes too slowly above 0.8 and very

fast below 0.5. Thus the switching surface was chosen as sk = [1,-1.66][ek 1, ek] in

order to get small error and short enough transient time as well.


Iterations Iterations
(a) (b)


60 100 150 200 250 300 0 50 100 150 200 250 300
Iterations Iterations
(c) (d)

Figure 5-22. Control of the liquid-level plant by the MQSMC varying the number of
controllers: (a)M= 1 (b)M= 16 (c)M= 36 (d)M= 144.

The control systems were tested for a typical square wave set-point of amplitude

0.5. Figure 5-22 compares the tracking performance of the reference signal, which is the

output of a first-order reference model with transfer function Gd (z) = 0.2z/(z 0.8) and









driven by the set-point, by the MQSMC varying the size of the map. The figure illustrates

how the control performance is affected by the number of QSM controllers. As the

number of controllers is increased the plant reaches the set-point with smaller steady-state

errors. Even if there seems, however, no big difference between Figure 5-22(c) and

Figure 5-22(d) the larger number of controllers might be working better in a noisy

environment. This will be discussed again later in this section. Additionally, a

comparison of this figure with Figure 5-20 displays the open-loop response and

establishes the efficacy of the proposed control scheme.

Table 5-7. Comparison of control performance for the liquid-level plant in noise-free
environment.

Methodology NRMSE

TDNNC (4:50:1) 5.7e-3
MIC (12x12) 1.4e-3
MPIDC (12x 12) 10.9e-3
MQSMC (12x 12) 5.0e-3

Table 5-8. Comparison of control performance for the liquid-level plant in the presence
of sensor noise: standard deviation of noise is 4.5e-2.

Standard deviation
Methodology of error
of error
TDNNC (4:50:1) 3.8e-2
MIC (12 x 12) 4.2e-2
MPIDC (12x 12) 10.2e-2
MQSMC (12x 12) 1.9e-2

Table 5-7 compares the tracking performance among controllers where MIC and

MQSMC outperformed TDNNC. However, even though MPIDC showed the worst

performance it showed the smallest steady-state errors among them. It only demonstrated






84


large errors in the transient phase since it was not fast enough to track the trajectory. In

addition, the controllers were tested on the same tracking trajectory regulating

disturbances whose standard deviation is 4.5e-2. The results are listed in Table 5-8.


a 50 100 160 200 250 30(
Iterations
(a)


Iterations
(b)


0 50 100 150 200 250 300 0 50 100 150 200 250 300
Iterations Iterations
(c) (d)

Figure 5-23. Control of the liquid level system with measurement noise by the
MQSMC with (a)M= 1 (b)M= 16 (c)M= 144 and (d) the TDNNC.

As in the previous result for controlling the first order plant, the MQSMC showed

the best performance in robustness against the measurement noise, and the MPIDC

showed the worst performance again due to the slow convergence. Figure 5-23 compares

the response among a nonlinear controller and the MQSMCs varying the number of the

controllers under a noisy environment. As we can see, the proposed MQSMC is capable

of rejecting measurement noise and following the trajectory with very little variations.


f









Also the MQSMC with M = 144 showed a more stable aspect than the smaller structure.

While a nonlinear controller, the TDNNC, showed relatively large chatters on the

trajectory even if it tracks the path, it also spent much more control effort to do the same

mission than the MQSMC did.


06 -- Plant output Plant output
S- Desired output Desired output









50 100 150 200 250 300 0 50 100 150 200 250 300
Iteration Iterations

Figure 5-24. Tracking an oscillatory reference signal of the liquid-level plant by the
TDNNC (left) and the MQSMC (right) in the presence of sensor noise.


>
VI
o0i
D



r,.


0



'2
0,




o


Incre


asing closed-loop variance


Figure 5-25. Performance assessment on a trajectory tracking under noisy environment.

Finally, we evaluated the tracking performance of an oscillatory reference signal by

the TDNNC and the MQSMC depicted in Figure 5-24. The reference signal to be

followed was 0.2(0.8 sin(0.15k + r / 2) +1.2 cos(0.1k)). We observed that the MQSMC

tracked the trajectory more smoothly than the TDNNC did. In summary, the proposed









MQSMC is the best possible multiple model based controller in a noisy environment for

both trajectory tracking and set-point tracking as illustrated in Figure 5-25. On the other

hand, the MPIDC is not suitable for a trajectory tracking problem that requires fast

transient response coping with measurement noise even if it is the best in set-point

tracking in the absence of noise.

5.3 Flight Vehicles

Flight vehicles, such as missiles and aircraft, are very complex systems that are

typically non-minimum phase and have aerodynamic coefficients which vary over a wide

dynamic range due to large Mach-altitude fluctuations [43,47]. Control of high-

performance low-cost UAV especially involves the problems of incomplete

measurements, external disturbances and modeling uncertainties. Nevertheless, the

autopilot for these vehicles is often required to achieve very stringent performance

objectives. Thus the proposed control algorithms are applied to missiles and UAV to

show the effectiveness in this section.

5.3.1 Missile Dynamics

We consider a benchmark missile model used widely in the literature [50,104].

This benchmark model can be formulated as

= + [93 (5.7)
9 2 0 q 9g4

and

g = [0.0001a2 0.0112|a 0.2010(2 -M/3)]M cos(a)
g2 = [0.0152a2 -1.3765a + 3.6001(-7+8M/3)]M2
g3 = -0.0403M cos(a)
g4 = -14.542M2










where a is the angle of attack (deg), q is the pitch rate (deg/s), M is the Mach number,

and 5 is the tail deflection angle (deg). We considered an operation range of M = 3 and

a e [-8, 8] deg. The model was formulated as Yk1 f(Yk Yk-1,UkUk k-1) where Yk is

the Angle Of Attack (AOA) to be controlled.






00I0 6o o 150 200 250 300 350 400



-00
DOE020 i------------------------------- Ji !- 10 -------------.----------







0 0 1 20 0 60 100 150 200 250 300 350 400
Number of PEs of one-side of a map Iterations
(a) (b)

Figure 5-26. Modeling performance using multiple models for the missile dynamics: (a)
Identification performance varying the network dimension (b) Identification
of a missile dynamics by 18 x18 multiple models.

The SOM was trained with the vector [Yk Yk 1 ] varying the size of the map and

tested with 400 newly generated data samples. Identification of the missile system with

18 x 18 multiple models, which was chosen as the optimal dimension of network as shown

in Figure 5-26(a), is depicted in Figure 5-26(b) where we observe that the models present

a very good approximation of the plant.

Table 5-9. Comparison of modeling performance for the missile system.


Methodology NRMSE

Multiple Models (18x 18) 6.1e-3

TDNN (4:35:1) 6.3e-3







88


The modeling performance using the designed multiple models were compared with that


by a TDNN, which was trained by one hidden layer with 35 PEs, in Table 5-9. The


proposed method showed marginally better performance than the nonlinear model did.


Based on 324 linear models and a nonlinear TDNN model with 35 PEs in the


hidden layer, we designed the multiple model-based controllers and the TDNN model


based controller. The optimal parameters selected for each controller were 30 hidden PEs


for TDNN controller, the desired pole locations of 0.3 and 0.7 for the closed loop


response by the PID controller, and c, = 1, c, = -1.85, qT = 0.51 for QSMC.








-6 --------------------- -6 ,-------------------
2 2
-4 -4




0 20 30 40 50 0 10 20 30 40 5 0
Time (sc) Time (sec)
(a) (b)


4 2



Z 0
4 2


2





4


0 10 20 30 40 50 600 10 20 30 40 50 60
Time ( ) TIme (se)

(c) (d)

Figure 5-27. Tracking various set-point reference signal by (a) TDNNC (b) MIC (c)
MPIDC and (d) MQSMC in the absence of measurement noise.