<%BANNER%>

Evaluating the Impacts of Advanced Driver Assistance Systems Using a Driving Simulator - an Exploratory Study

Permanent Link: http://ufdc.ufl.edu/UFE0042636/00001

Material Information

Title: Evaluating the Impacts of Advanced Driver Assistance Systems Using a Driving Simulator - an Exploratory Study
Physical Description: 1 online resource (140 p.)
Language: english
Creator: Martin, Barbara
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adaptive, adas, car, changing, control, cruise, driving, following, lane, simulator
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre: Civil Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: EVALUATING THE IMPACTS OF ADVANCED DRIVER ASSISTANCE SYSTEMS USING A DRIVING SIMULATOR - AN EXPLORATORY STUDY There is an increasing number of vehicle technologies being developed and deployed during the past few years. Some of these technologies can take control over specific functions of the vehicle, while others provide warnings to assist drivers in a variety of driving tasks. These are called Advanced Driver Assistance Systems (ADAS), and were designed mainly to improve roadway safety and provide comfort to drivers. There is evidence that these systems may be able to result in traffic operational improvements and congestion mitigation, but a limited amount of research has been conducted to assess these potential impacts. This thesis evaluated traffic operations under two types of ADAS in a driving simulator (STISIM Drive) environment. Two systems, which are more likely to affect traffic operations, were evaluated: Adaptive Cruise Control (ACC) and Lane Change Assist (LCA). A specific route was created in the driving simulator, which consisted of an arterial section followed by a freeway. This route was driven twice by drivers: first without the systems and secondly using the two ADAS. There were a total of 25 participants. During the test, performance measures such as speed, lane change maneuvers and headway with the front vehicle were collected. The analysis compared the data obtained without the systems to those obtained while the systems were used. Results showed changes in driving behavior due to the systems and a potential positive impact of these technologies in traffic operations.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Barbara Martin.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Elefteriadou, Ageliki L.
Local: Co-adviser: Washburn, Scott S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042636:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042636/00001

Material Information

Title: Evaluating the Impacts of Advanced Driver Assistance Systems Using a Driving Simulator - an Exploratory Study
Physical Description: 1 online resource (140 p.)
Language: english
Creator: Martin, Barbara
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adaptive, adas, car, changing, control, cruise, driving, following, lane, simulator
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre: Civil Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: EVALUATING THE IMPACTS OF ADVANCED DRIVER ASSISTANCE SYSTEMS USING A DRIVING SIMULATOR - AN EXPLORATORY STUDY There is an increasing number of vehicle technologies being developed and deployed during the past few years. Some of these technologies can take control over specific functions of the vehicle, while others provide warnings to assist drivers in a variety of driving tasks. These are called Advanced Driver Assistance Systems (ADAS), and were designed mainly to improve roadway safety and provide comfort to drivers. There is evidence that these systems may be able to result in traffic operational improvements and congestion mitigation, but a limited amount of research has been conducted to assess these potential impacts. This thesis evaluated traffic operations under two types of ADAS in a driving simulator (STISIM Drive) environment. Two systems, which are more likely to affect traffic operations, were evaluated: Adaptive Cruise Control (ACC) and Lane Change Assist (LCA). A specific route was created in the driving simulator, which consisted of an arterial section followed by a freeway. This route was driven twice by drivers: first without the systems and secondly using the two ADAS. There were a total of 25 participants. During the test, performance measures such as speed, lane change maneuvers and headway with the front vehicle were collected. The analysis compared the data obtained without the systems to those obtained while the systems were used. Results showed changes in driving behavior due to the systems and a potential positive impact of these technologies in traffic operations.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Barbara Martin.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Elefteriadou, Ageliki L.
Local: Co-adviser: Washburn, Scott S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042636:00001


This item has the following downloads:


Full Text

PAGE 1

1 EVALUATING THE IMPACTS OF ADVANCED DRIVER ASSISTANCE SYSTEMS USING A DRIVING SIMULATOR AN EXPL ORA TORY STUDY By B RBARA BARQUETA MARTIN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Brbara Barqueta Martin

PAGE 3

3 To m y parents, sister and boyfriend

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my graduate advisor, Dr. Lily Elefteriadou for her guidance throughout this thesis and her valuable support throughout my years of study at the University of Florida. I would also like to thank the remaining members of the committee, Dr. Scott Washburn, Mr. William Sampson and Dr. She rrilene Classen, for their assistance and their advices. I would like to thank my boyfriend Daniel for his assistance and constant support, motivating me through the toughest times and being my companion all along this journey. Above all, I am grateful t o my parents, Marilda and Reinaldo for their love, support and dedication that enabled me to be who I am

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Background ................................ ................................ ................................ ............. 12 Problem Statement ................................ ................................ ................................ 12 Research Objectives ................................ ................................ ............................... 13 Document Organization ................................ ................................ .......................... 13 2 LITERATURE REVIEW ................................ ................................ .......................... 14 Advanced Driver Assi stance Systems (ADAS) ................................ ....................... 14 Adaptive Cruise Control ................................ ................................ .................... 15 Lane Change Assist ................................ ................................ ......................... 17 Driving Simulators ................................ ................................ ................................ ... 18 General Use ................................ ................................ ................................ ..... 18 Testing ADAS ................................ ................................ ................................ ... 20 I mpact of ADAS on Traffic ................................ ................................ ...................... 21 Literature Findings ................................ ................................ ................................ .. 22 3 METHODOLOGY ................................ ................................ ................................ ... 25 Description of the Driving Simulator ................................ ................................ ........ 25 Implementation of ACC and Lane Change Assist ................................ ................... 27 Driving Scenarios ................................ ................................ ................................ .... 30 User Acceptance Questionnaire ................................ ................................ ............. 32 Data Collection ................................ ................................ ................................ ....... 33 Data Analysis ................................ ................................ ................................ .......... 33 4 DATA COLLECTION ................................ ................................ .............................. 38 Participant Selection ................................ ................................ ............................... 38 Recruitment and Prescreening ................................ ................................ ......... 38 Characteristics ................................ ................................ ................................ .. 39 Data Collection Procedures ................................ ................................ .................... 41 Final Data Types and Formats ................................ ................................ ................ 43

PAGE 6

6 5 DATA ANALYSIS ................................ ................................ ................................ .... 50 Traffic Impact of the Technologies ................................ ................................ .......... 50 Impact of ACC ................................ ................................ ................................ .. 51 Impact of LCA ................................ ................................ ................................ ... 56 Number of lane changes ................................ ................................ ............ 56 Minimum distances during accepted gaps ................................ ................. 58 Average distances during accepted gaps ................................ .................. 61 General Impact of Both Systems ................................ ................................ ...... 63 User Acceptance of the Technologies ................................ ................................ .... 65 Summary and Findings ................................ ................................ ........................... 66 6 CONCLUSIONS ................................ ................................ ................................ ..... 80 Research Summary ................................ ................................ ................................ 80 Research Conclusions ................................ ................................ ............................ 80 Futu re Research and Recommendations ................................ ............................... 83 APPENDIX A ................................ ......................... 85 B SCENARIO FILES ................................ ................................ ................................ 109 C QUESTIONNAIRES ................................ ................................ .............................. 123 LIST OF REFERENCES ................................ ................................ ............................. 136 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 140

PAGE 7

7 LIST OF TABLES Table page 2 1 Assessment of traffic efficiency of advanced driver assistance systems ............ 24 3 1 User acceptance questionnaire ................................ ................................ .......... 37 4 1 ................................ ................................ ................ 45 4 2 TSS score for gender groups ................................ ................................ .............. 45 4 3 TSS score for age groups ................................ ................................ ................... 46 4 4 Driver behavior type results ................................ ................................ ................ 47 4 5 Demographic s characteristics by driver behavior type ................................ ....... 48 4 6 Raw data for subject 25 with the systems ................................ .......................... 49 5 1 Car following data for subject 1 with the ACC ................................ .................... 67 5 2 Car following aggregate data for each driver, without and with ACC .................. 68 5 3 Detailed car following analysis for subjects 19 and 25 ................................ ....... 69 5 4 Average headway (ft) and standard deviation for gender groups ....................... 71 5 5 Average headway (ft) and stan dard deviation for age groups ............................ 71 5 6 Average headway (ft) and standard deviation for driver behavior types ............. 71 5 7 Number of lane change maneuvers per subject, without and with LCA .............. 72 5 8 Number of lane change maneuvers for gender groups ................................ ...... 72 5 9 Number of la ne change maneuvers by age groups ................................ ............ 73 5 10 Number of lane change maneuvers for driver behavior groups .......................... 73 5 11 Minimum distances ( ft) for lane changes performed on the arterial .................... 74 5 12 Minimum distances (ft) for lane changes performed on the freeway .................. 7 5 5 13 Average of minimum distances (ft) for arterial for gender groups ....................... 75 5 14 Average of minimum distances (ft) for freeway for gender groups ..................... 76 5 15 Average of minimum distances (ft) for arterial for age groups ............................ 76

PAGE 8

8 5 16 Average of minimum distances (ft) for freeway for age groups .......................... 76 5 17 Average of minimum distances (ft) for arterial for driver behavior groups .......... 76 5 18 Average of minimum distances (ft) for freeway for driver behavior groups ......... 76 5 19 Average distances (ft) for lane changes performed on the arterial ..................... 77 5 20 Average of distances (ft) for gender groups ................................ ....................... 77 5 21 Average of distances (ft) for age groups ................................ ............................. 77 5 22 Average of distances (ft) for driver behavior groups ................................ ........... 78 5 23 Average speed (mi/hr) for each subject ................................ .............................. 78 5 24 Average of average speed (mi/hr) for gender groups ................................ ......... 79 5 25 Average of average speed (mi/hr) for age groups ................................ .............. 79 5 26 Average of average speed (mi/hr) for driver behavior groups ............................ 79 5 27 Average user acceptance for all subjects ................................ ........................... 79

PAGE 9

9 LIST OF FIGURES Figure page 3 1 Simulator field of view (including rear view mirror dis play) and computer workstation ................................ ................................ ................................ ......... 35 3 2 Junction box that distributes computerized signals to steering wheel, accelerator, brake and retrieves turn signal indications. ................................ ..... 35 3 3 Emplo ying speed to control headway. ................................ ................................ 35 3 4 Parameters for headway control ................................ ................................ ......... 36 3 5 Unsafe zone for Lane Change Assist ................................ ................................ 36 3 6 Arterial section used in the second scenario ................................ ...................... 37 4 1 Data collection steps ................................ ................................ .......................... 48 5 1 Distribution of headway without and with ACC ................................ ................... 69 5 2 Vehicle trajectory for a car following situation (Subject 14) without ACC ........... 70 5 3 Vehicle trajectory for a car following situation (Subject 14) with ACC ................ 70

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EVALUATING THE IMPACTS OF ADVANCED DRIVER ASSISTANCE SYSTEMS USING A DRIVING SIMULATOR AN EXPLORATORY S TUDY By B rbara Barqueta Martin December 2010 Chair: Ageliki Ele fteriadou Cochair: Scott Washburn Major: Civil Engineering There is an increasing number of vehicle technologies being developed and deployed during the past few years. Some of these technologies can take control over specific functions of the vehicle wh ile others provide warnings to assist drivers in a variety of driving tasks. The se are called Advanced Driver Assistance Systems (ADAS), and were designed mainly to improve roadway safety and provide comfort to drivers. There is evidence that these systems may be able to result in traffic operational improvements and congestion mitigation, but a limited amount of research has been conducted to assess these potential impacts. This thesis evaluate d traffic operation s under two types of ADAS in a driving simu lator ( STISIM Drive ) environment. T wo systems which are more likely to affect traffic operations were evaluated : Adaptive Cruise Control (ACC) and Lane Change Assist ( LCA ). A specific route was created in the driving simulator which consisted of an arte rial section followed by a freeway. This route was driven twice by drivers: first without the systems and secondly using the two ADAS. There were a total of 25 participants. During the test, performance measures such as speed, lane change

PAGE 11

11 maneuvers and hea dway with the front vehicle were collected. The analysis compared the data obtained without the systems to those obtained while the systems were used. Results showed changes in driving behavior due to the systems and a potential positive impact of the se te chnologies in traffic operations

PAGE 12

12 CHAPTER 1 INTRODUCTION Background Advanced Driver Assistance Systems (ADAS) are electronic devices installed in vehicle s to assist drivers in tasks such as lane changing, merging and speed control by providing warnings or even taking control of the vehicle. These systems have shown promise in the improvement of road safety, as proven in several tests (Golias et al. 2002; Touran et al. 1999) An important aspect, brought recently by research, is that these systems may also be able to result in traffic improvements and congestion mitigation. Some papers alre ady show positive effects of one particular system evaluated yet. Driving simulators have proven to be a great tool used in exper iments to assess various changes in the dr iving environment, vehicle control, and driver behavior. They provide a controlled environment, necessary to expose the driver to situations without putting the participant in danger. Problem Statement Existing literature regarding ADAS is focused primari ly on safety effects. Also, research related to traffic operations typically consider each system separately and do not consider their combined effects. Finally, the literature has not evaluat ed specific traffic measures in depth considering different driv gender and aggressiveness With the constant increase of these technologies on the market, traffic impacts represent an important subject to be considered.

PAGE 13

13 Research Objectives This thesis evaluate s selected traffic operat ional performance measures under two types of ADAS in a driving simulator environment. This research is a first step towards a broader evaluation of ADAS impacts on traffic, by addressing two systems integrated in one vehicle. The user acceptance of the sy stems will also be evaluated through a simple questionnaire. More specifically the objectives of this study are: To implement two types of ADAS, Adaptive Cruise Control and Lane Change Assist, in the STISIM driving simulator, with scenarios that include ar terial and freeway driving. To collect data from human subjects in two types of driving environments (with and without the ADAS) in the driving simulator and to obtain information about the experience involving these new technologies and the factors that could affect the behavior of the drivers through a questionnaire. To compare the trajectories of vehicles with and without instrumentation, focusing on headways with the front vehicle average speed and completed lane change maneuvers. To evaluate driving behavior changes due to the systems in different driver types. Document Organization Chapter 2 presents a literature review on A dvanced D river A ssistance S ystems, focusing on the ones used in this project ( Adaptive Cruise Control and Lane Change Assist ) t he use of driving simulators to test these systems, and the known impacts of these technologies. Chapter 3 describes the methodology used to conduct the experiments and procedures to analyze the data Chapter 4 presents the description of data collection, and Chapter 5 presents the data analysis The final chapter presents a summary of the research, along with conclusions and recommendations.

PAGE 14

14 CHAPTER 2 LITERATURE REVIEW This chapter summa rizes past research related to Advanced D river Assistance S ystems and the use of driving simulators. The f irst section describes the advanced driver assistance systems (ADAS) that will be examined in the thesis : Adaptive Cruise Control an d Lane Change Assist. The second section summarizes previous research that used driving simul ators to assess these systems. The third part presents p revious research that evaluate d the impact of ADAS on traffic. Th e chapter concludes with a summary of the literature findings. Advanced Driver Assistance Systems (ADAS) ADAS refer to electronic devices for the support of drivers in performing various driving tasks (such as merging, speed control or lane keeping), by providing real time advice, instruction and warnings or even intervening in the vehicle control and maneuvering tasks. The use of t hese systems is expected to improve road safety, increase road capacity and also attenuate environmental load in traffic (Golias et al. 2002, Wiethoff et al. 2002) Improvement of road safety is related to the al leviation of the driving task, decreasing accident consequences. Brookhuis et al. (2001) discussed the behavio reaction time and situation awareness. Experiments for safety assessment of these technologies have been extensively conducted over the past years An increase in road capacity is relate d to the changes in traffic dynamics, reflected in measures such as mean driving speed and optimized headway distances (Golias et al. 2002) This research focus es on the autonomous systems, inst ead of the cooperative systems. The first ones use on board equipment to assess the surrounding

PAGE 15

15 environment, and can be implemented on the current road infrastructure; the cooperative approach r el ies on the communication between all vehicles or road (Piao and McDonald 2 008) Since the autonomous systems are already available on the market, they are t he ones tested in this thesis. The choice of which systems would be implemen ted in this research was based o n Table 2.1, edited from Golias et al. (2002) which summarizes the traffic efficiency expected by each of the available systems. Platooning, adaptive cruise contro l and lane change and merge collision avoidance are the only vehicle support systems that have a potential to impact traffic efficiency. Although platooning has a probable impact on both speed and headway adjustment, its application areas are usually restr icted to highway network sections with a reduced speed limit (usually up to 53 mi/hr ), and it was no t tested in this experiment. The two systems ev aluated in this thesis are the A daptive C ruise C ontrol and L ane Change A ssist. Adaptive Cruise Control Adapti ve Cruise Control (ACC) is an upgrade of the conventional cruise control. It requires the driver input ing into the system a maximum speed and a time headway. When the ACC is being used, it maintains the set maximum speed when there are no vehicles ahead, a nd adjusts the speed automatically to follow other vehicles at a constant time gap previously defined by the driver when a slower vehicle appears in front ACC is a longitudinal control system and uses a sensor to detect the presence of the front vehicle, measuring the relative speed and distance to it (Guvenc and Kural 2006) In this system, the throttle and brake are controlled by the computer, while only the

PAGE 16

16 steering is controlled manually. When the sensor detects a fr ont vehicle, the computer sends the appropriate command to the throttle and brake. (Ioannou and Chien 1993) ACC systems are now available on a wide range of passenger vehicles and they are the most common form of ADAS since there is no need for a special infrastructure or inter vehicle communication. (Auckland et al. 2008) The experiments with ACC include reliability assessment, driver acceptance, safety issues, consequences on traffic flow and even environmental effects. Fancher and Ervin (1998) describe the field operational test with ten passenger cars incorporating the system, evaluating how the ACC techn ology would work under real operating conditions. Fancher et al. (2001) show through an operational test of ACC the limitation of the acceleration and deceleration levels, and Hoedemaeker (2000) indicates an approval coming from both low and high speed drivers. The global impact of ACC on safety of highways was studied by Touran et al. (1999) concluding that the system significantly reduces the probability of collision with the ACC and front vehicle. Another safety issue is the need f rom the driver to understand the capabilities of the system, including braking and sensor limitations, so the driver is able to intervene in situations that exceed ACC capabilities, and preventing them to rely on the system inappropriately (Seppelt and John D. Lee 2007) Stanton et al. (1997) observed that the ACC may result in a reduced mental workload, which may reduce the level of attention, affecting the ability of the driver to maintain awareness of the system.

PAGE 17

17 Experiments by Hoedemaeker (2000) demonstrate that driver behavior with ACC leads to positive effects on traffic efficiency, since it reduces sp eed variability and differences in driving behavior, harmonizing traffic. Also the short headways to increase roadway capacity are accepted very well by drivers Liang and Peng (2000) show that ACC vehicles can help improve the average speed and reduce the average acceleration levels of mixed traffic, leading to higher traffic flow rates. Regarding environmental eff ects, Bose and Ioannou (2003) indicate that with 10% of vehicles hav ing ACC, the reduction in air pollution ca n be up to 60% arguing that the smooth response of ACC vehicles designed for human factor considerations filters out traffic disturbances. Lane Change Assist Lane Change Assist (also called Collision Avoidance or Warning System) is a lateral control syste m that monitors the position of vehicles traveling in close proximity in adjacent lanes and in the same direction, warning the driver when it is unsafe to change lanes or merge into a line of traffic. These in vehicle electronics systems are rearward looki ng, a nd mainly radar based. The objective is to assist drivers who are intentionally attempting to chang e lanes by detecting vehicles in the driver's blind spot Most of the systems use the information from tracking vehicles just to warn the driver when their position and/or speed makes the planned lane change/merge maneuver unsafe, but more sophisticated systems may include speed and steering control intervention for enhanced collision avoidance (Golias et al. 2002) Primarily, this t echnology focuses on safety improvement. Lane change maneuvers can be a dangerous situation when the driver overlooks other vehicles or

PAGE 18

18 underestimates their speed (Ruder et al. 2002) These systems ensure that lateral separation between vehicles in adjacent lanes are always maintained and may therefore have significant positive impacts in the reduction of traffic accidents (Mazzae et al. 1998) An extensive safety research study sponsored by the U.S. Department of Transportation (U.S. DOT), the Integrated Vehicle Based Safety Systems (IVBS S) includes the lane change warning as one of the systems in an integrated safety system for light vehicle platform s (Ference 2006) Golias et al. (2002) analyzed the traffic impacts observing that the optimized lane change and merging capabili ties of these systems may lead to significant traffic efficiency gains, related to better headways, although they are expected to have little or no impact on The traffic impact of this technology needs to be studied in depth, since no research analyzing this topic was found. Most of the experiments with this system are focused on the technology related to vehicle detection, in an attempt to minimize false alarms. Driving Simulator s General Use Driving simulators are used eff icient ly for vehicle system development, human factor s stud ies and other purposes because they reproduce actual driving conditions in a safe and controlled environment. Driving simulators use computer based technology to create a virtual reality that gives the driver an on board impression of a real driving experience The simulator predicts vehicle motion caused by driver input and feeding back corresponding visual, motion and audio cues to the driver (Lee et al. 1998) The use of research simulators has many advantages over similar on road driving experiments. For example, t hey can be configured to simulate a variety of research

PAGE 19

19 problems and create different scenarios to match the requirements of a particular experiment. With simulators it is possible to control experimental conditions making every driver drive the same testing situation where systematic variation in road, vehicle or traffic situations conditions are difficult to achieve in the real world. The performance measures are also easy to be obtained and there is no risk to the drivers or other road users (Blana 1996) However, there are some possible disadvantages that may include the lack of validation as a predictor of on road driving performance, the accurate replication of physical sensations and, the onset of simulator sickness (Godley et al. 2002) Research using I MAP (Institute for Mobillity, Activity and Participation) simu lator STISIM Drive Model 500 W already found relative and absolute validity Shechtman et al. (2009) findings support validity to the STISIM simulator, observing that the same trends when negotiating turns in the simulator can be transferred to the road when testing conditions are the same. Shechtman et al. (2010) also found relative validity in a group of healthy licensed drivers between driving tasks on road and i n the simulator. O ther considerable disadvantage s may occ ur while some subjects driv e the simulator, and experi ence adverse effects, known as simulator adaptation response or simulator sickness Common symptoms can be grouped into nausea, oculomotor discomfort, and disorientation (Mourant and Thattacherry 2000) In some cases this condition is a critical limitation that disable s people from completing the simulated driving assessment

PAGE 20

20 The degree of the effects experience d by the onset of simulator sickness can be evaluated through the Simulator Sickness Questionnaire (SSQ) (Kennedy et al. 1993) This questionnaire co nstitutes 16 symptoms on side effects across oculomotor, disorientation and nausea domains and the participant rates them using a 4 point Likert scale ( 0=absent, 1=slight, 2=moderate, 3=severe). A Total Severity Score (TSS) is comput ed, indicating severity of SS symptoms felt by the subject where zero indicates no sickness and a high TSS s evere symptoms Testing ADAS In the last few years, many experiments to assess different types of ADAS used driving simulator experiments, demonstr ating it as a suitable tool to be used in these studies. Riener and Ferscha (2008) used a driving simulator to evaluate a vehicle control system based on vibro tactile feedback against the common used interaction modalities vision and soun d of driver assistance systems. Their experiments demonstrate that this approach is a viable alternative to on road user st udies. Most of the experiments assessing ADAS using a drivin g simulator are related to ACC. Lin et al. (2009) conduc ted a study on a bus driving simulator to investigate the effects of time while reclaiming control from ACC in a car following scenario of emergency brake by the lead vehicle. Lee et al. (2006) from ACC to manual control when warned with different alerts, by comparing headway maintenance performance for drivers with or without ACC during brake events in a driving s imulator. Park et al. (2006) used a driving simulator to create virtual ACC failure

PAGE 21

21 distance between cars, to investigate the effect of behavioral adaptation on driv ers driving and control ability. The behavioral influences of ACC were studied by Saad (2004) showing considerable chang es in speed, safety margins to the front vehicle and frequency of lane change maneuvers, but he points out the need to account for long term adaptation. Regarding other types of ADAS, a serial steering assist controller to help avoid road departure was des cribed by Chen and Ulsoy (2006) using driving simulator experiments to provide validation for the technology I mpact of ADAS on Traffic The research developed for evaluation of these assistance systems is not just focused on safety, but also on their potential impact on traffic. The majority of the papers use models and simulation to study the effects of ACC vehicl es. Kerner (2003) showed that ACC vehicles can improve the efficiency of the systems by suppressing wide moving jams and therefore traffic flow can be stabilized and harmonized in a synchronized mode. But on the downside, at certain parameters, ACC vehicles can induce congestion at bottlenecks. Through detailed simulations of a section of the German autobahn, Treiber and Helbing (2001) reported that nearly all congestion would be eliminated if 20% of the vehicles were equipped w ith ACC system. Even with 10%, travel times due traffic jams were reduced by more than 80%. The results from Bose and Ioannou (2003) through simulation of mixed traffic, demonstrated that ACC vehicles help smooth traffic flow by filtering the response of rapidly accelerating lead vehicles. The same simulations were used by Ioannou and Stefanovic (2005) to analyze disturbances that may arise due to

PAGE 22

22 lane changes, demonstrating that ACC vehicle response c an attenuate the perturbation due to a cut in vehicle A hybrid model was used by Yuan et al. (2009) to investigate the traffic breakdown probability from free flow to congested flow, and the transition probability from synchronized flow to jams. ACC vehicles were demonstrated to enhance the traffic stability of synchronized flows. Other experiments show the negative i mpact of ACC, like Vander Werf et al. (2002) w ho shows that increases above 60% of ACC vehicles can lead to a modest loss of highway capacity, based on average time gaps longer than those for manual vehicles. Literature Findings According to the literature, ADAS represent a new set of technolog ies with great potential to improv e safety and road capacity by chang ing the way people drive. Previous research has shown that these systems have a behavioral impact on drivers, but most of them do not show the effects of these impacts on traffic flow. Although studies with ACC already showed positive effects on traffic, t he majority of the experiments with ADAS are linked to safety issues. Lane Change assist systems have not yet been evaluated in depth regarding their potential traffic impacts The literature also shows a lack of experiments integrating some of the systems and evaluations that consider different driver types In an attempt of being more realistic, the well consolidated systems should be tested together in the same vehicle, so the evaluation can account with the effects of all the systems, including the poss ible interaction between them.

PAGE 23

23 According to the literature driving simulators are a suitable tool extensively used to test and evaluate new technologies as well as driver behavior aspects. Therefore, it is appropriate to use a driving simulator for the p urposes of this research.

PAGE 24

24 Table 2 1. Assessment of traffic efficiency of advanced driver assistance systems Traffic efficiency VEHICLE Speed adjustment Headway adjustment general vehicle control automatic stop and go L L Platooning H H spee d control L L adaptive cruise control H H collision avoidance road and lane departure collision avoidance L L lane change and merge collision avoidance L H rear end collision avoidance L L obstacle and pedestrian detection L L intersection colli sion avoidance L L vehicle monitoring Tachograph L L alerting systems L L vehicle diagnostics L L H: high, important impact; L: low, limited or insignificant impact

PAGE 25

25 CHAPTER 3 METHODOLOGY This chapter presents the methodology used in the experiment to evaluate the two ADAS. The chapter first provides a description of t he driving simulator at the University of Florida, and then summarizes the steps through which it was prep ared for the given experiment. These consist of the implementation of ACC and Lane Change Assist algorithms and the creation of the scenarios to be displayed In order to assess user acceptance of the two systems, a methodology found in the literature was used and is described in the fourth part of the chapter. The data collection a nd the data analysis pla n are briefly presented in the fifth and sixth parts of this chapter respectively Description of the Driving Simulator The Driving Simulator is located at the Institute for Mobility, Activity an Participation at the University of F lorida. It provides a large forward field of view of 180 degrees and displays virtual objects behind the car (real image side and rear view mirrors) making the en tire scene computer generated. Figure 3 1 shows a picture of the simulator field of view. Thr ee 3 x 6 foot flat screens connected with scenes provided by three high intensity projectors (Sanyo, 2000 ANSI Lumens) make the large field of view possible. In order to better represent the forward scene it would be preferable to have a continuous curvil inear surface, but subjects indicate that they notice no partitioning of the three separate segments after a few minutes of practice n.d.) The o verall contrast can be altered to simulate reduced visibility associated with heavy rainfalls or fog.

PAGE 26

26 The simulator is integrated with a vehicle ( 1997 Dodge Neon ) where the driver operates normal accelerator, brake, and signaliz ing and steering controls with the corresponding visual scene responding accordingly. Longitudinal and lateral movement allows the driver to speed up or slow down, come to a halt, and steer sideways including making lane changes and chang ing of direction at intersections. All these changes are controlled by software that interfaces with a junction box under the hood of the vehicle, shown in Figure 3 2. The simulator is built on a computerized platform developed by the Systems Technology Inc. (STI) of Haw thorne, CA. The specific configuration is the STISIM Drive Model 500W produced by STI. The vehicle and tire model runs on a dedicated processor that is linked to the simulation via a network. It operates at fast update rates necessary to provide high fid elity simulation of the vehicle dynamic responses and provides proper steering force/feel feedback. Road feel is also captured via a low frequency audio woofer and amplifier providing engine, transmission and road noise at varying intensities and frequenc ies, all of this simulated by the STISIM Drive software. The software also has the capability of substituting standard sounds provided with the in the driving scenario. A control area and a workstation are situated to the rear of the vehicle overlooking the driver, vehicle and viewing screens. From this area the researcher can communicate with the subject by a two way communication, maintained via speakers and microphon es in the vehicle and at the workstation. The three visual screens are

PAGE 27

27 duplicated at the workstation and a fourth control monitor allows the experimenter to set parameters for each trial and to monitor data being collected. The recording software permits the acquisition of up to 40 vehicle, driver and simulation parameters. Specific data recorded depends upon the driving scenario being used and the assessment goals Implementation of ACC and Lane Change Assist The systems were implemented in the driving simulator through the STISIM Drive Open Module Programming, by creating a custom Open Module DLL. The code was programmed by computer science student s in Visual Basic 6.0 and is available in Appendix A The ACC algorithm was based on Fancher et al. (1998) employing an approach that uses speed to control headway. The main conceptual features of this algorithm are that it maintains the maximum speed set up by the driver if no impeding traffic prevails, it adjusts th e speed to maintain the desired headway with slower traffic, and it autonomously switches back and forth between the two operational modes. Figure 3 3 illustrates this concept. The algorithm is based on the following functions and operations that character ize the system: The system will never achieve a speed higher than that selected by the driver If the driver brakes, the system disengages and does not automatically reengage thereafter If the driver accelerat es, the system automatically reengages thereafter with the previous set up parameters Preceding vehicles at 525 ft or less are the only ones considered.

PAGE 28

28 The first step of the algorithm is to search for the first vehicle in front of the subject vehicle and in the same lane and compute the distance between them. Followi n g that, the algorithm i s divided in a simple case, when there is no slower traffic ahead (beyond 525 ft), and an advanced case, when the headway is adjusted according to the lead vehicle (less than 525 ft away). maintain the maximum speed set up by the driver (V max ). The acceleration (a) has to be computed every time step to adjust the control, and is calculated accordi ng to Mezny et al. (2009) : (3 1) Where: V max is the maximum speed set up by the driver, T is a calibrated parameter to control the reaction time of the algorithm. This parameter allows for a smooth transition of speeds during ACC control, and it could result in extremely high acceleration values when the difference between V max and V is very high. Therefore, this parameter can be interpreted as the react ion time of the system. F or this simulator after a series of tests, the selected value of T wa s 5 The maximum acceleration defined was 10ft/sec 2 and the maximum deceleration was 20 ft/sec 2 (based on the simulator capabilities). The advanced case require s the desired time headway set up by the driver (T h in seconds), which is a control system parameter. The desired gap distance (R h in feet) is a linear function of lead vehicle speed (V l ):

PAGE 29

29 (3 2) R h is considered as the desired gap distance (from the ACC vehicle to the lead vehicle) towards which the controller attempts to converge. Figure 3 4 illustrates these parameters. The parameter R h is used by the algorithm to compute the speed command (V c ), which is the desired speed for the ACC vehicle to achieve the desired gap (R h ) in a smooth manner: ( 3 3) Where V c is the speed command (the maximum is V max ) V l is the speed of the leading vehicle, R is the c urrent gap distance, R h is the desired gap, T 0 determines the closing rate and as suggested by Fanch er et al. (1998) equals 11 sec. The desired acceleration is calculated according to equation (3 1), substituting V max for V c According to Mezny et al. (2009) to limit the number of repetitions and avoid the algorithm reacting to minor changes, the acceleration is updated only if either of these conditions is met: R differs from R h by more than 1% (if R equals 199ft and Rh is 200ft, the difference is 0.5% and the condition is not met), V l differs from V c by more than 5% (if V l equals 100 ft/sec and V c is 98 ft/sec, the difference is 2% and the condition is not met), The difference between V l and V c is g reater than the difference between V c and V (If V l (100 ft/sec) minus V c (98 ft/sec), that is 2 ft/sec, then V needs to at least 96.9 ft/sec for this condition to be met).

PAGE 30

30 The Lane Change Assist algorithm was based on LeBlanc et al. (2008) The algorithm implemented used a simplified rationale, adjusted to the simulator characteristics. It provides a warning sound w hen the subject vehicle is changing lanes and has the turn signal on, and at least one vehicle is within the unsafe zone (Figure 3 5). The warning sound beeps continuously when the turn signal is on until there is no vehicle inside the zone. If the subject vehicle attempts a lane change maneuver without turning on the signal indicating the target lane, the system is not triggered ; therefore there is no warning sound. Driving Scenarios A driving scenario constitutes the environment that the driver will navig ate through while interacting with different events at specific times. It defines the roadway design, the surrounding traffic, background, signs, pedestrians, and many other events related to a driving environment. All subjects drove the simulator 4 times, including a total of 3 scenarios, the same for all drivers. The first was an acclimation scenario so drivers could adapt to the differences between a simulator and a real car. It starts at low speeds, with signalized intersections to provide stop and go c onditions. Consequently, drivers face an increase on speed limits and have to make turns. Along this scenario there is no traffic in the same direction as the subject vehicle. The duration was at most 12 minutes (scenario files have a fixed highway length, performance was not collected during the acclimation scenario The second scenario participants drove was the main one c reated and it was used twice to collect vehicle performance data, both without and with the ADAS This

PAGE 31

31 scenario was created so drivers would surely use the systems and experience various situations to analyze their impact. It star ts on an arterial with th ree 12 feet lanes in each direction, and a speed limit of 40 mi/hr. There is heavy traffic surrounding the vehicle, and most of the time surrounding vehicles are fairly slower than the speed limit, creating situations w here the drivers would be most likely to change lanes. This tactic was used because the simulated vehicles cannot change lanes as a function of traffic and do not vary their speed according to the scenario files (vehicles are assigned to a lane and a certa in speed). They do vary their speed (speed changes are not controlled by parameters in the scenario file). The subjects have to go through signalized intersections and m ake turns. The entire arterial section is presented in Figure 3 6. The arrows show the path subjects were instructed to drive through. The freeway section is 19,500 feet long, with two 12 feet lanes in each direction separated by a median. There are 3 curv es along the freeway, and the speed limit along the entire freeway is 65 mi/hr. Traffic along the freeway is heavy and vehicles brake and accelerate according to traffic ahead, and can reach speeds as low as 40 mi/hr. This freeway section was created to ma ximize car following opportunities. There are no merge/diverge situations because the simulator cannot handle them accurately (there is a possibility to create these facilities, but the simulator does not recognize them properly). The entire scenario (arte rial and freeway) could be completed in 9 to 11

PAGE 32

32 This scenario was driven by participants first without any advanced system, and all relevant data were collected. Afterward each participant drove the third s cenario, consisting of a straight roadway section and low traffic levels, to familiarize themselves with the technologies tested (3 minutes total). The last scenario to drive was the second one with the add ition of the assistance systems, where data were a lso collected. The second scenario file used on this experiment to collect data is presented in Appendix B. User Acceptance Questionnaire A simple procedure for the assessment of acceptance of new automotive technologies is described in Van Der Laan et al. (1997) The questionnaire they developed provides a five point scale to assess nine attributes of a technology, and has been used to measure driver acceptance for a large number of research products Since the attribute set is sufficiently broad, the procedure can be used directly to compare the acceptance of d ifferent systems. For e ach item the scale is anchored by two polar adjectives, and drivers are asked to rate their perception of the system by marking a box between the two poles. The scale is scored from 2 to +2, with positive numbers corresponding to values closer to the positive adjectives (being +2 the closest to the positive adjective) and vice versa. Table 3 1 shows the nine items in the questionnaire (Van Der Laan et al. 1997) After a principal components analysis, Van Der Laan et al. (1997) suggests that the scale can reliably be reduced to two subscales: usefulness (items 1, 3, 5, 7 and 9) and satis faction (items 2, 4, 6 a nd 8). T he first step in the analy sis of the responses is to determine whether the items within each subscale correl ate well with each other. To determine this correlation, t he authors recommend obtaining a

PAGE 33

33 least 0.65 as a scale reli ability analysis. Cronbach's alpha is a measure of internal consistency, that is, how closely related a set of items are as a group It determines the internal consistency or average correlation of items in a survey instrument to measure its reliability (Reynaldo and Santos, 1999) After this criterion is met, the component scores for each subject will be aver aged to get usefulness and satisfaction scores for each subject. The two scores averaged across subjects represent the overall perception associated with the systems, where positive numbers lead to positive perception and vice versa. This questionnaire wa s conducted after the driving experience to measure acceptance of both ACC and Lane Change Assist, as well as the combination of the systems Data Collection Data w ere collected during the two times participants drove the second scenario (one without the t ested systems and the other using them) The data include d information related to the subject vehicle, such as speed, acceleration, lane position and longitudinal position, and also information related to each vehicle that is in the roadway display, such a s lateral position, longitudinal position with respect to the subject vehicle and difference in speed relative to the subject vehicle This dataset w as used to analyze the impact of the advanced systems. A more detailed description of the data collection p rocess is presented in Chapter 4. Data Analysis The final step of the methodology includes the data analysis procedure. The purpose of this effort is to observe how driver behavior changes when the two assistance systems are used during driving experience Driving characteristics were

PAGE 34

34 observed during particular driving tasks that could affect traffic dynamics. To analyze whether ACC could lead to driving behavior changes, car following situations without and with the system were statistically compared using space headways. Driving characteristics with the use of the LCA were evaluated using the number of completed lane change maneuvers and accepted gaps during lane changing. Average speeds were used to analyze the overall impact of both systems. The complete analysis of the data is presented in Chapter 5

PAGE 35

35 Figure 3 1. S imulator field of view (including rear view mirror display) and computer workstation (Source: http://driving.phhp.ufl.edu/virtual) Figure 3 2. Junction box that distributes computerized signals to steering wheel, accelerator, brake and retrieves turn signal indications (Source: http://driving.phhp.ufl.edu/virtual) Figure 3 3. Employing speed to control headway (Source: Fancher et al. (1998) )

PAGE 36

36 Figure 3 4. Parameters for headway control (Modified from Fancher et al. (1998) ) Figure 3 5. Unsafe zone for Lane Change Assist (both sides are symmetric) V l

PAGE 37

37 Figure 3 6. Arterial section used in the second scenario Table 3 1. Use r acceptance questionnaire 1 useful useless 2 pleasant unpleasant 3 bad good 4 nice annoying 5 effective superfluous 6 irritating likeable 7 assisting worthless 8 undesirable desirable 9 raising alertness sleep inducing

PAGE 38

38 CHAPTER 4 DATA COLLECTION This chapter presents the data collection effort. Obtaining behavioral data requires the involvement of the Institutional Review Board (IRB) of the University of Florida. The Behavioral/Non Medical department of IRB is responsible for reviewing and monitoring all research studies that involve human subjects, such as this one. Thus, all appropriate forms, questionnaires, as well as the research methodology, were approved by the IRB prior to the data collection. The qu estionnaires are provided in Appendix C The data collection occurred during the month s of August and September 2010, at Oak Hammock at the University of Florida, where the simulator is located. The process involving the selection of participants, data col lection procedures and final data types and formats are described below. Participant Selection Recruitment and Prescreening The advertisement for recruitment was posted at public locations including several supermarkets and University of Florida campus, at http://gainesville.craigslist.org/ and sent to graduate students. A prescreening procedure was designed to obtain age, gender, driving experience and health conditions. Respondents could download the que stionnaire from the website ( https://sites.google.com/site/drivingsimulatorproject/ ) or send an email to the researcher, and respond offline through email or mail. Any person with a val id driver license, with at least 3 years of experience and no health problems was considered a qualified candidate. Age and gender w ere also considered to ensure a diverse group of participants. Following the questionnaire, each participant considered for the experiment was assessed for cognitive function through

PAGE 39

39 the Mini Mental State Examination (MMSE) (Folstein et al. 1975) This is a brief 30 point questionnaire test that is used to screen for cognitive impairment Part of it was administered through telephone, assessing 22 items, and the rest was performed in person before the testing. A minimum of 17 out of 22 was necessary for the first part to schedule an appointment, and the minimum in general had to be 2 6 over the 30 points to participate in the experiment (participants with no cognitive impairment) Characteristics aphics (age and gender), their Total Severity Score (TSS) in relation to simulator sickness and a driver behavior type analysis. Age, gender and driver behavior type groups will be used in the data analysis chapter. A total of 32 subjects participated in the experiment, but 7 of them did not complete the test due to excessive simulator sickness. The demographics of the participants, the maximum TSS (where zero indicates no sickness) and whether the subject completed the test or not are presented in Table 4 1. As shown in this table there were a total of 9 female and 16 male participants that completed the test. With respect to age distribution 11 participants were between 25 and 30 years old, while 7 were between 31 and 40, and 7 between 41 and 60 years old. The goal of having participants distributed equally among gender and age was not accomplished as a consequence of the high propensity of sickness in older and female drivers A simple analysis was conducted after the experiments to observe how SSQ sc ores correlated to age and gender. Table 4 2 shows the maximum TSS averaged by gender group.

PAGE 40

40 A comparison of means through a t test (at 95% confidence level) confirms that male participants had lower scores than females, and consequently felt less sick. Th e comparison between age groups is presented in Table 4 3. Regarding age groups, the only significant difference was observed between 40 60 and 25 29 years old participants, where the oldest one had greater SSQ scores tha n the youngest at 90% confidence le vel. This confirms that females and older adults are more susceptible to SS (Allen et al. 2006) In order to analyze how the variability in drive r behavior affected the impact of the two technologies tested, driver behavior types were identified after the experiments. Three types of driver behavior were distinguished : aggressive, average and conservative behavior. The assessment was based on two cr iteria (AAA Foundation for Traffic Safety, 2009) : (i) observed speed under free flowing and not car following conditions, and (ii) number of discretionary lane changes. Given the design of the scenarios (i.e., heavy traffic throughout the driving scenario ), these criteria were adapted to fit the data. Therefore, number of lane changes on the arte rial and freeway, and how many mile/hour over the speed limit participants achieved were used in this assessment. As such, participants that performed at least 10 lane changes on the arterial and 4 on the freeway or achieved a speed over 10 mi/hr the speed limit in one facility and more than 5 in the other were considered aggressive. Participants that met either 2 of the criteria: (i) performed between 7 and 9 lane changes on the arterial, (ii) 2 or 3 on the freeway, (iii) achieved speed over 10 mi/hr the s peed limit on the freeway, (iv) achieved a speed over 10 mi/hr the speed limit on the arterial ; were considered to be average. Participants that did not achieve speeds over 10 mi/hr the speed limit

PAGE 41

41 (arterial or freeway) and performed less than 7 lane chang es on the arterial and 2 or less on the freeway were considered conservative. Table 4 4 presents the results of the driver behavior analysis for participants, along with the parameters they chose for ACC (maximum speed and time headway) and the background survey responses Although choices for ACC parameters can agree with some driver types (for example, conservative drivers chose the lowest maximum speeds ), the fact that participants are first time users of the systems needs to be taken in consideration M ore familiarity with the systems may lead these choices to correspond more accurately with the driver behavior types. The background survey responses (questionnaire presented in Appendix C) to some degree come in agreement with the results of the driving b ehavior analysis. Most of the drivers characterized by themselves and others as somewhat conservative were considered as conservative drivers in this analysis The ones that considered themselves somewhat aggressive are divided in aggressive and average dr ivers. The inconsistencies presented may be due to the fact that participants respond to the questionnaire by comparing themselves with their peers, which does not provide objective responses (Kondyli 2009) 4 5. Although the sample is not equally divided in gender and age groups, it was possible to achieve the goal of having a diverse sample in terms of aggressiveness. Data Collection Procedures After taking the prescreening questionnaire and part of the MMSE on the phone, license, a nd have a light snack before the testing session, but not a heavy, large or

PAGE 42

42 greasy meal. The session could last as little as 1 hour and as long as 2 hours, and the compensation was set a t $50 per participant. Once in the simulator, a set of steps was under taken as illustrated in Figure 4 1. A description for each one of these steps is provided. Step 1 Upon arrival and before starting the experiment, a check in procedure was undertaken as follows: 1) sign the informed consent form and give a copy to the s ubject 2) complete the background survey form, and 3) provide a photocopy of explains the entire experiment, followed by the agreement of the driver. At this point, the resear ch discussed the early signs of simulator sickness and told the subject that as soon as they notice any symptom, they need to stop the simulation and take a break. They were also informed that they would be kept for some minutes after the simulation if the y were not feeling fine. The background survey questionnaire provided information related to their driving habits and also to their perceived degree of aggressiveness. Step 2 Participants were screened for simulator sickness using the Simulator Sickness Questionnaire SSQ (Kennedy et al. 1993) to identify subjects that could experience simulator sickness. If there was a sign that the subject wa s prone to is a FDA approved medical device that provides a relief from nausea due to motion sickness by gentle electrical stimulation of the nerves in the wrist. Step 3 The y were taken to the vehicle, and the researcher explained exactly how the simulator works. The participant was then subjected to an acclimation period in the simulator, beginning with a less complex visual representation of the road environment, with progr essive increases in complexity. Step 4 After the acclimation period, participants completed the SSQ again, identifying those that experienced any symptom of simulator sickness. If this happen ed the steps 5a through 8a were followed: o Step 5a The part icipant took a break, until he/she fe lt well again; o Step 6a The driver answered the SSQ; o Step 7a and then dr ove another acclimation period. o Step 8a The SSQ was answered again. If the symptom persist ed the simulation was discontinued (Step 9) and the participant was dismissed (Step 10). If no symptom was observed, the experiment continue d with Step 5b.

PAGE 43

43 Step 5b T he participant was asked to drive the first scenario as he/she would normally do on the road This first scenario was driven without the as sistance systems and it took about 9 11 minutes. Step 6b In this step the researcher demonstrate d to the subject how the two assistance systems work by letting them drive a 3 minute scenario with the systems. They were then asked what was the maximum sp eed they would achieve in a freeway with a speed limit of 65 mi/hr, and what was their preferable time headway Step 7b The participant drove the second scenario using the two assistance systems. Step 8b After finishing the driving experiment, the p articipant answered a questionnaire related to user acceptance of the new systems and the SSQ, so he/she did not leave the place feeling any sick. Step 10 The participant was paid and dismissed. Final Data Types and Formats The simulator provided an out put file for each time the participant drove the second scenario (without and with the system). This file contains information of the subject vehicle and the other vehicles on the display every 0.5 second. For the subject vehicle, the output provides speed (mi/hr), acceleration (ft/sec 2 ), longitudinal and lateral position (ft). For the other vehicles, the file contains vehicle ID, difference in speed (ft/sec) and longitudinal position (ft) in relation to the subject vehicle and lateral position (ft). An exa mple of this raw data is presented in Table 4 6. This table shows information for only one vehicle (Vehicle ID 29) in addition to the obtained contain information on all vehicles in the display for every time step. For the case wher e the participants used the systems, two other files were generated with information on when the systems were being used (when ACC was on, and when LCA made the warning sound). This information was aggregated in the output files. From these files, detailed information on car following situations and lane change

PAGE 44

44 maneuvers were extracted along with average speeds. The analysis of this dataset is provided in Chapter 5.

PAGE 45

45 Table 4 Subj ID Gender Age TSS Completion 1 F 38 0.00 Y 2 M 53 0.00 Y 3 M 29 0.00 Y 4 M 57 0.00 Y 5 M 31 7.48 Y 6 F 46 14.96 Y 7 M 26 0.00 Y 8 M 36 0.00 Y 9 F 31 26.18 Y 10 M 25 0.00 Y 11 F 28 7.48 Y 12 M 32 0.00 Y 13 F 29 18.70 N 14 M 28 0.00 Y 15 M 38 0.00 Y 16 F 29 3.74 Y 17 F 28 18.70 N 1 8 F 55 26.18 N 19 F 26 0.00 Y 20 F 50 14.96 Y 21 M 47 14.96 Y 22 F 42 0.00 Y 23 M 59 3.74 N 24 F 38 18.70 N 25 M 44 0.00 Y 26 M 25 0.00 Y 27 F 28 14.96 N 28 F 29 0.00 Y 29 M 54 18.70 N 30 M 25 0.00 Y 31 M 26 0.00 Y 32 M 40 18.70 Y *F female M male, Y yes, N no. Table 4 2. TSS score for gender groups n Average TSS Standard deviation F 14 11.75 9.71 M 18 3.53 6.72

PAGE 46

46 Table 4 3. TSS score for age groups Age n Average TSS Standard deviation 25 29 14 4.54 7.36 30 39 7 7.48 10.80 40 6 0 11 10.20 9.62

PAGE 47

47 Table 4 4. Driver behavior type results Experiment observations Arterial Freeway Background Survey Responses ACC parameters Subj ID Over 40 (mi/hr) #LC Over 65 (mi/hr) # LC Driver Type Speed (mi/hr) Lane Change Aggressiveness by themselves Aggressiveness by others Vmax (mi/hr) Th (sec) 1 11.34 5 7.09 3 Aggressive 70 75 Very often Somewhat aggr. Somewhat aggr. 75 2 2 10.20 10 6.65 0 Aggressive 70 75 Sometimes Somewhat cons. Somewhat cons. 75 2 12 14.46 10 22.36 1 Aggressive 70 75 Very often Somewhat aggr. Somewhat aggr. 75 2 15 6.57 6 14.74 2 Aggressive 70 75 Very often Somewhat cons. Somewhat cons. 70 2 23 4.91 11 3.04 4 Aggressive 70 75 Very often Somewhat aggr. Somewhat aggr. 75 2 24 12.65 10 16 .33 2 Aggressive 70 75 Sometimes Somewhat aggr. Somewhat aggr. 75 2 25 12.08 14 10.19 4 Aggressive 70 75 Very often Somewhat cons. Somewhat cons. 75 2 4 3.20 7 5.01 3 Average 70 75 Very often Somewhat aggr. Somewhat cons. 65 2 5 5.52 10 3.87 3 Av erage 70 75 Sometimes Somewhat cons. Somewhat cons. 72 2 7 2.91 7 12.94 2 Average 70 75 Very often Somewhat cons. Somewhat aggr. 80 2 9 8.99 8 0.62 2 Average 70 75 Very often Somewhat aggr. Very cons. 70 2 16 13.38 6 1.92 2 Average 70 75 Very ofte n Somewhat aggr. Somewhat aggr. 70 2 19 5.26 9 2.05 5 Average 70 75 Very often Somewhat aggr. Somewhat aggr. 70 1.5 20 9.38 6 7.45 2 Average 70 75 Very often Somewhat aggr. Somewhat aggr. 75 1.5 22 8.39 5 8.39 0 Average 70 75 Very often Somewhat a ggr. Somewhat aggr. 75 1.5 3 2.22 6 7.29 1 Conservative 65 70 Very often Very cons. Very cons. 65 2 6 0.55 4 6.09 1 Conservative 70 75 Sometimes Somewhat cons. Somewhat cons. 75 2 8 0.58 6 0.92 1 Conservative 70 75 Very often Somewhat cons. Some what cons. 70 2 10 3.18 4 5.34 1 Conservative 70 75 Sometimes Somewhat cons. Somewhat cons. 70 2 11 0.67 4 1.08 0 Conservative 70 75 Sometimes Somewhat cons. Somewhat cons. 73 2 13 0.59 5 2.08 0 Conservative 70 75 Sometimes Somewhat cons. Somewha t aggr. 70 2 14 3.75 6 2.36 0 Conservative 70 75 Very often Somewhat cons. Somewhat cons. 70 2 17 1.48 6 0.04 1 Conservative 70 75 Sometimes Somewhat cons. Somewhat aggr. 65 2 18 0.88 5 5.11 1 Conservative 70 75 Very often Very cons. Very cons. 75 2 21 2.09 5 2.62 2 Conservative 70 75 Sometimes Somewhat aggr. Somewhat aggr. 69 2

PAGE 48

48 Table 4 5. Demographics characteristics by driver behavior type Driver type Male Female Average age Aggressive 5 (71%) 2 (29%) 34.7 Average 4 (50%) 4 (50%) 35 .8 Conservative 7 (70%) 3 (30%) 35.2 All 16 (64%) 9 (36%) 35.2 Figure 4 1 Data collection steps

PAGE 49

49 Table 4 6. Raw data for subject 25 with the systems Time (sec) Speed (m i/hr ) Accel (ft/sec2) Long dist (ft) Lateral position (ft) Min range (ft) V ehicle ID Diff. in speed (ft/sec) Diff. long position (ft) Lateral position (ft) 0.0 0.0 0.5 0.0 6 43.5 29.0 58.7 817.1 30 0.5 0.1 0.5 0.0 6 63.5 29.0 45.7 791.3 30 1.0 0.1 0.5 0.0 6 83.5 29.0 32.8 771.9 30 1.5 1.7 6.3 0.7 6 102.8 29.0 27.2 758.1 3 0 2.0 3.9 6.6 2.9 6 120.6 29.0 28.0 744.3 30 2.5 6.1 6.7 6.7 6 136.8 29.0 28.7 730.2 30 3.0 8.4 6.7 12.2 6 151.3 29.0 29.4 715.7 30 3.5 10.7 6.7 19.4 6 146.3 29.0 30.1 700.9 30 4.0 12.9 6.6 28.2 6 123.3 29.0 30.8 685.7 30 4.5 15.2 6.6 38.6 6 102 .0 29.0 31.5 670.2 30 5.0 17.4 6.5 50.7 6 82.5 29.0 30.2 654.8 30 5.5 19.6 6.5 64.4 6 64.7 29.0 27.8 640.4 30 6.0 21.8 6.4 79.8 6 48.8 29.0 25.3 627.3 30 6.5 24.0 6.3 96.7 6 35.1 29.0 22.6 615.4 30 7.0 25.5 2.8 115.1 6 24.1 29.0 20.9 604.8 30 7 .5 26.4 2.7 134.2 6 17.5 29.0 19.8 594.7 30 8.0 27.3 2.7 153.9 6 17.0 29.0 18.7 585.1 30 8.5 28.2 2.6 174.3 6 17.3 29.0 17.6 576.1 30

PAGE 50

50 CHAPTER 5 DATA ANALYSIS The first part of this chapter summarizes the findings regarding the traffic impact of th e two technologies. This impact was analyzed through three different evaluations, one focusing on the impact of ACC, the other on the impact of LCA and the last one on the general impact of both systems. The second section of the chapter applies the method ology described in chapter 3 (Methodology User Acceptance Questionnaire) to evaluate the user acceptance of the technologies. This procedure evaluates how drivers perceive the two technologies by assessing the usefulness and satisfaction of each system. The chapter concludes with a summary of the findings regarding the impact of the two systems. Traffic Impact of the Technologies This analysis evaluate s the impacts of the technologies on traffic operational performance measures I t focuses on (a) the imp act of the ACC as expressed by the space headway s of the subject vehicle with the front vehicle (b) the impact of LCA as expressed by the number of lane chang ing maneuvers completed by the subject and the characteristics of the gaps accepted, and (c) the general impact of both systems as expressed by the speed of the subject vehicle. The se performance measures obtained from the two driving experiences (with and without the systems) were statistically compared at 95% confidence level, to evaluate c hanges in the availability of the systems. Means, standard deviations, and distributions were statistically compared to assess differences in the two scenarios driven by the same driver, and also between groups of drivers (gender, age and a ggressiveness)

PAGE 51

51 Impact of ACC The Adaptive Cruise Control was used by drivers only along the simulated freeway (not on the arterial), because the system works during high speeds only. Prior to the beginning of the simulation, drivers were asked to choose a maximum speed that the vehicle would automatically achieve in case there was no slower traffic ahead, and a time headway that the vehicle would maintain when it encountered a slower lead vehicle. The subjects were asked to base their decision on their dri ving experience and regular practices, and to facilitate the perception of time headway, three options were presented : 2.0, 1.5 and 1.0 sec. The impact of ACC was evaluated through car following situations along the freeway segment. These situations were s elected such that the subject vehicle followed the same lead vehicle for a minimum of 20 seconds. For the scenario with the systems, the ACC had to be on during the selected car following situations. O n average, participants used the ACC on the freeway 77% of time This percentage represents the time the system was on The system can be turned on by pushing the button, and letting th e algorithm control the vehicle and off by using the brake pedal T hese conditions were repeated several times along the freew ay). For both scenarios, the first 1,200 ft of the freeway were not considered as this area constitutes the transition al area from arterial to freeway. Depending on the lane change maneuvers performed, the number of car following situations recorded for each participant varied between 1 and 5 per participant per scenario. The space headway recorded every 0.5 second was averaged for each car following situation, and the respective standard deviation was also computed. Table 5 1 shows a typical car followin g situation using ACC with the subject vehicle having higher

PAGE 52

52 speeds until it reache d the lead vehicle. At that time it starts adjusting its speed to maintain the headway selected by the driver (subject 1). In many cases the number of car following situati ons was more than one per participant per scenario The average space headway and standard deviation of headways for these cases w ere calculated aggregating all car following situations per participant. Table 5 2 presents this information for each subject, with and without the systems As shown, the average space headway was reduced for most, but not all subjects when ACC was used. The average difference of space headway between without and with the system was 42.5 ft. A paired samples difference t test was perf ormed to determine whether the space headway per participant during car following w as higher without the system than with it. The t test indicates that the use of ACC resulted in significantly smaller headways. To compare the standard deviation with a nd without the systems, F tests were performed for each driver. From this analysis, two s ubjects ( 7 and 16 ) did not exhibit difference s in the variability of headway s during car following ( calculated F was lower than the critical value ) F or 7 subjects (2, 10, 11, 17, 20, 23 and 24) the standard deviation was actually higher with the ACC, but for most of the subjects (1 6 of the 25) the standard deviation with ACC was smaller than without it Table 5 3 presents a detailed car following analysis for subjects 19 and 25, showing the average space headway and standard deviation for each car following situation performed with and without ACC These two drivers performed three or more car following situations during the freeway segment, and exhibited a great varian ce

PAGE 53

53 across average space headway when the system was not available in agreement with the previous analysis. Subject 25 presents lower standard deviations per car following situation when the system is being used, although the average space headway per car following situation seems higher with ACC than without it. This table illustrates the fact that there is variability under different situations even for the same driver. Figure 5 1 shows the distribution of headways for all car following situations withou t and with the ACC. These graphs show th at when the ACC wa s used, the s pace headway distribution shifted significantly toward the left of the graph, and th e frequency of longer headways wa s minimized Figures 5 2 and 5 3 show vehicle trajectories for a giv en subject (subject 14), without and with ACC. These trajectories illustrate the difference between the two cases for the same driver. Comparing them, it can be noticed that the subject vehicle with ACC decelerated in a smoother way as it approache d the le ad vehicle In contrast, the subject vehicle without the ACC was more impacted by the deceleration of the lead vehicle, and the space headway between the two wa s not as s mooth as with the ACC. When the lead vehicle br aked the subject vehicle without ACC g ot closer to it and t ook longer to adjust its headway. W hen the lead vehicle wa s accelerating, the sub ject vehicle without ACC also took longer to get closer to the lead vehicle than with ACC. Next, the impact of ACC on car following situations is evaluate d by gender, age, and driver behavior groups. Gender group s Th is sub section evaluates how gender affects the differences in outcomes between the two tests (with and without the systems), and also whether there is a significant difference in driving perf ormance between these two groups. Table 5 4

PAGE 54

54 shows the average space headway and overall standard deviation for female and male groups, for both driving tests The overall standard deviation is obtained by aggregating car following situations from all drive rs in the specific group. A paired samples difference t test was performed to determine whether the use of the system shortened the space headway during car following compared to when the system was not present. In the last column of the table, it can be observed that this difference was approximately 40 ft for the male group and 47 ft for the females The t test indicated that this difference is statistically significant for the female group, but not for the male one Comparison of the standard deviation with and without the systems (last column of Table 5 4 ) shows that both groups had significantly less variab ility when using the systems (F equals 2.06 for male and 1.24 for female F critical equals 1.2 1 ). The difference in space headway between female an d male drivers is not statistically significant in both tests (with or without the systems). Without the ACC, male participants had a higher standard deviation in headways compar ed to female participants However, with the ACC, both groups exhibited the sa me variability in headways Therefore, ACC is significantly helping male drivers maintain their space headway more stable during car following Along the freeway section, female participants used the ACC 75% of the time in average, while male ones used it 79% of the time. Age group s The second subdivision grouped subjects in 2 categories: between 25 and 30 years old, and between 31 and 60 years old This categorization was used because the number of younger participants was bigger (less susceptible to simu lator sickness) and because previous analysis with two older groups (31 to 40 and 41 to 60

PAGE 55

55 years old) did not exhibit additional information In results, these two were grouped in drivers between 31 to 60 years old. Table 5 5 presents the results for each group and differences between scenarios driven without and with the system and between the two groups. O lder and younger drivers reacted differently to ACC As shown in the table, f or younger participants, the standard deviation was the same with the ACC and without it. In contrast, for older ones, the system significantly contributed to smaller headways (about 100 ft) and less variability of headways. Comparison of the two groups shows that without the system older drivers presented higher standard deviat ion of headways than younger ones, while with the system this was inverted having older drivers varying their headways less than younger ones. Although younger drivers used the system for more time than older ones (younger used it 81% of time, older 75%), the older ones were the ones positively affected by the system. Driver behavior type s The last subdivision uses driver behavior types to group participants. Table 5 6 summarizes the results for each group. The conservative group was the only one that con tributed significantly to shorter headways with the ACC ( space headways w ere about 105 ft s horter with the system ). Regarding the variability of headways, all 3 groups showed smaller standard deviations of headways with the system comparing to without it C onservative drivers compared to aggressive and average ones, had significant ly longer headway s without the ACC, but not with the system (headway among the groups became equivalent) For standard deviation, conservative drivers presented the highest value s, while average

PAGE 56

56 ones presented the lowest one (both with and without the ACC). Interestingly, aggressive driver s had a higher variability of headways than average drivers in both tests. These drivers used the system 67% of the time, while average and cons ervative ones used it 81% of the time. Summary of ACC impact. Along the simulated freeway, participants in general had smaller space headways with the lead vehicle and less headway variability during car following situations. The analysis by group showed t hat female drivers, older and conservative ones were the ones that reduced their headways the most when using the ACC. All groups drove more smoothly, and reduced the variability of their headways. This finding is consistent with the literature findings wh ich indicate the ACC system contributes to more stable traffic. The use of the ACC also contributed to eliminating differences in space headways between driver behavior types Impact of LCA The Lane Change Assist was available along the entire scenario (bo th freeway and arterial), providing a warning sound if the driver was attempting an unsafe lane change maneuver while the indicator to the target lane was on. The system was activated ( it beeped) on average 4.2 times per test. Number of l ane c hanges The fi rst measure use d to evaluate the impact of LCA is the number of lane change maneuvers performed by each driver without and with the system, separately on the arterial and the freeway, as well as the total number of lane changes for the entire scenario. Th e simulator output provided the lateral position of the center of the subject vehicle every time step. The beginning of a lane change maneuver was considered as

PAGE 57

57 the time step when the center of the vehicle was more than 3 feet away from the center of the 1 2 feet lane (same rationale applied to the end of the maneuver). This deviation was determined by observing the lane changing patterns in the driving simulator (no specifics found in the literature) Table 5 7 presents the number of lane change maneuvers f or each driver. As shown, in most cases lane changes increased when the LCA was used. S tatistical comparison showed that the average number of lane change maneuvers increased significantly for the arterial and the entire scenario (combined arterial and fre eway) when LCA was present The increase in lane changes for the freeway section was not found to be significant Along the freeway section drivers were also using the ACC systems, which could have given them some comfort related to braking and accelerati ng tasks. It is possible that subjects did not feel the need to change lanes ally react to the traffic ahead, since the ACC was performing such tasks. Next, t his dataset was subdivided in age, gender and driver behavior groups to evaluate the LCA impact on each subgroup. A paired samples difference t test was performed to compare average number of lane changes. Gender groups The summary of the analysis by gender group is presented in Table 5 8 Statistical comparison shows that male participants perform ed significantly more lane changes along the entire route when using the system (1.9 more lane changes) T he difference between the number of lane change maneuvers with and without the system for female drivers on the arterial and in total was 1.9 and 1.1 which was not found to be statistically significant. In contrast, the t test shows that female

PAGE 58

58 participants performed less lane changes on the freeway with the LCA Also on the freeway, m ale participants performed more lane change s when using the system in comparison to females. Age groups The data aggregated by age is presented in Table 5 9 Statistical comparison shows that a significant increase in number of lane changes on the arterial (1.4) and for the entire scenario (1.7) o ccurred for the older group when using the system. The differences for t he younger category between with and without the system presented in the first row of the table were not significant showing that these drivers were not affected by the technology Th e number of lane changes from the younger group did not differ statistically from the older one in any of the tests (with and without). Driver behavior types. Data on driver behavior groups is presented in Table 5 10 Although there was a n increase in numb er of lane changes for aggressive and average drivers when using t he LCA on the arterial and entire test, these differences were not statistically significant. However, the system contributed to significant more lane changes along the scenario for conserva tive drivers. Compari son between average and aggressive drivers shows no difference in number of lane changes, whether with or without the system. But when comparing these two groups with the conservative group, it was statistically shown that they perform ed more lane changes when the system was not present. The presence of the systems eliminated this difference for the freeway section (all three groups performed statistically the same number of lane changes along this facility). Minimum d istances during ac cepted gaps Another aspect that was used to evaluate the impact of LCA is the distances during accepted gaps and how they change with the assistance system. In this section,

PAGE 59

59 the minimum distances compared refer to the ones between the subject vehicle and t he adjacent vehicles (on the target lane), both with the vehicle in front and back of the subject vehicle. These measures were compared with and without the technology. The analysis is performed for the arterial and freeway portions separately For each se gment, the minimum of the total gap and the individual distances to both vehicles (front and back of the subject vehicle) on the adjacent target lane were analyzed. Tables 5 1 1 and 5 1 2 show the minimum distances between all the lane changes performed per subject, for the arterial and freeway portions respectively. As shown in Table 5 1 1 drivers on the arterial accepted total gaps significantly shorter when using the LCA Without the system, the average minimum gap accepted was 199 ft, while with it this d istance dropped to 150 ft. The other distances (with the back and front vehicle on the target lane) did not show significant differences when the system was available compared to without LCA On the freeway, the average of minimum distances accepted was si gnificantly shorter when considering the back and front vehicle separately, having a decrease of a pproximately 30 and 170 ft respectively when the system was available. Although the total gaps without the system were 120 ft bigger than with it, this differ ence was not statistic ally significant. In order to explore the impact of LCA following analysis groups the participants in gender, age and driver behavior types. A paired samples difference t test was performed t o compare minimum distances across groups and tests (with and without the system).

PAGE 60

60 Gender groups. Table 5 1 3 shows the average of minimum distances for each gender group on the arterial, while Table 5 1 4 presents this information for the freeway. On the ar terial, female drivers had a significant decrease on the accepted distances with the back vehicle and as a total during lane change when the system was available (65.5 and 110.4 ft respectively). Female group had accepted distance with the front vehicle as the only one significantly bigger than the male groups when the system was not available. With the LCA this difference no longer existed. On the freeway, the difference of 44.1 ft between distances with the back vehicle without and with the LCA for the m ale group, and of 223.9 with the front vehicle is statistically significant. The female group exhibited no significant difference for accepted distance. Although some differences between groups (last row of Table 5 1 4 ) seem large, they are not statisticall y significant. Age groups. The data is presented for each age group on Tables 5 1 5 and 5 1 6 for arterial and freeway respectively. On the arterial, the younger group was the only one affected by the LCA This group accepted significant shorter distances w ith the vehicle behind (difference of 43.7 ft comparing without and with the system test) and total gaps (58.1 ft difference). Comparing the younger with the older group, there was no significant difference. Although the differences between without and wit h the system test o n the freeway seem of a great magnitude, they were not statistically significant. The comparison between younger and older group was not significant with or without the LCA Driver behavior types. Tables 5 1 7 and 5 1 8 present the averag e of minimum distances grouped for driver behavior types, on the arterial and freeway. On the arterial,

PAGE 61

61 the system had no contribution to significant smaller headways for any specific driver behavior type Comparison b etween these groups showed that aggres sive and average drivers accept ed distances with the back vehicle significantly shorter than the conserva tive ones, regardle s s of whether the system was available. On the freeway, conservative drivers accepted shorter distances with the front vehicle with the system compared to without it A verage and aggressive drivers presented significantly shorter accepted distances when changing lane compared to conservative drivers without the LCA The presence of the system eliminated these differences. Average dista nces during accepted gaps The next analysis used the overall average distances during accepted gaps for each participant. Table 5 1 9 shows the average accepted distances between all the lane changes performed per subject. Analysis with average accepted dis tances shows that in general, drivers accepted significantly s horter distances with the LCA than without it. The distance with the vehicle 60 ft, and the total gap a ccepted dropped 112 ft. The impact of LCA is presented below with an analysis divided in gender, age and driver behavior groups. Gender groups Average distances for each gender group is presented in Table 5 20 For b oth female and male participants, the average of total gap and distance with the front vehicle accepted during lane change were significantly shorter with the LCA than without it. The female group notably dropped their accepted gap from 524 ft to 359 ft. C omparing the female group to the male one, no significant difference was found.

PAGE 62

62 Age groups The data is presented for each age group on Table 5 2 1 For the two age groups, the average of total gap and distance with the front vehicle accepted were significa ntly shorter with the LCA Both groups dropped their accepted total gaps by approximately 1 1 0 ft. Compari son between younger and older group showed no significant difference. Driver behavior types. Table 5 2 2 presents the average accepted distances grouped in driver behavior types. Conservative drivers accepted shorter average total gaps and distance with the front vehicle, with the LCA than without it. The difference in total accepted gap for this group between without and with the system tests was 172.5 f t. For average drivers, only the average distance with the front vehicle was significantly shorter (difference of 57.8 ft) when using the LCA Aggressive drivers did not change their gap acceptance characteristics when using the system. A verage and aggres sive drivers exhibited shorter distances with the back vehicle compare d to conservative ones when the system was not available. For total gaps, aggressive drivers accepted shorter distances than the conservative ones without the LCA When the system was us ed, these differences no longer appeared. Summary of LCA impact The impact of LCA was analyzed through the number of completed lane change maneuvers and with minimum and average accepted distances during a lane change maneuver for each driver. The number of lane change maneuvers, in general, was significantly higher with the system along the arterial and total route indicating that drivers may feel more comfortable to perform lane changing when the system is available. For the freeway section the number o f lane changes did not change significantly, possibly because the ACC was also being used. The male,

PAGE 63

63 older and conservative groups were the ones that most ly changed their behavior when using the system. For the minimum accepted distances on the arterial du ring lane changing, total gaps and distance with the vehicle behind were shorter with the system. Female and y ounger drivers presented shorter minimum distance s on the arterial when using the system, while male and conservative ones presented shorter minim um distances on the freeway. For average accepted distances during completed lane change maneuvers, drivers in general accepted shorter distances when using LCA For both age and gender groups, the average accepted distances dropped. Regarding driver beha vior, c onservative drivers were the most affected by the system, accepting shorter distances General Impact of Both Systems The presence of the two assistance systems may have an impact on driver behavior that is not directly related to the functionality of them (ACC may impact not just car following situations and LCA may impact more than the lane change maneuvers). In this section, average speed was used as a parameter to evaluate the overall i mpact of these two technologies when used together Table 5 2 3 provides the average speed (mi/hr) for each driver and for each test Data are provided for each analysis segment : arterial, freeway and total scenario. ADAS were higher tha n those without it. S tatistical comparison shows that average speed increased under two of all three cases (arterial and tot al scenario) when the systems were available. For the freeway section, there was a tendency to higher speeds with the systems that a pproached statistical significance. Additionally an

PAGE 64

64 analysis comparing maximum and minimum speeds was performed for the freeway section showing no statistical significance. Next, this data set is subdivided in to groups to analyze the impact of ADAS as a fun ction of different driver characteristics. Gender groups The data summary for each group is presented on Table 5 2 4 Female drivers increased their speed on the arterial section (0.9 mi/hr higher with the systems) and on the entire scenario (0.5 mi/hr hig her). For male participants the same happened on the arterial (0.7 mi/hr higher). The male group achieved higher speeds on the arterial and as a total compared to the female group when there was no ADAS. When the systems were used, only the speed on the fr eeway was significantly hi gher for male participants compared to female ones Age groups Table 5 2 5 shows the aggregated data by age group. On the arterial, both groups achieved significantly higher speeds with the systems compared to without them (averag e of 0.8 mi/hr higher) The total average speed only increased for t he young er group (0.6 mi/hr) Comparing the older to the younger group, no significant difference was found. Driver behavior groups Table 5 2 6 summarizes the data for different driver beh avior groups. A s observed before, aggressive drivers did not change their behavior due to the technologies. For the speed comparison, only a verage drivers had significant ly higher speeds on the arterial and total scenario (1.2 mi/hr and 0.6 mi/hr increase respectively). Without the systems, aggressive drivers had higher speeds than the average ones, who had higher speeds than the conservatives for all three cases

PAGE 65

65 (arterial, freeway and total). When the systems were available, the only significant differenc e was between aggressive and conservative drivers on the arterial. Summary of general impact The systems tested resulted in an increase in speeds for both segments (arterial, freeway) and for the entire scenario. Age groups (female and male) and gender gr oups (younger and older) were positively affected by the systems (higher speeds). For driver behavior types, only the average drivers significantly increased their speed when using the systems. User Acceptance of the Technologies Another important way to a nalyze these technologies is to evaluate how drivers accept them. This can determine whether drivers would be willing to use these systems in the long term. T his section applies the methodology presented in Chapter 3 to determine the usefulness of the syst ems and how satisfied the drivers were in relation to their use. Table 5 2 7 shows the data for this analysis, divided in ACC, LCA and the combination of the two technologies. The s cale for usefulness and satisfaction ha s a maximum of 2 and minimum of 2 T he estimation of of the methodology, and it needs to be at least 0.65 ( the criteri on was met for all cases). In general, subjects thought the systems were useful, and were satisfied more with ACC and the combined technolog ies, than with just the LCA This was a consequence of the sound chosen to warn drivers, classified as annoying by most subjects Further analysis of these results showed that age, gender or driver behavior types makes no difference in how drivers perceive the usefulness and satisfaction of the two ADAS systems and the combination of them. All drivers, independent of their characteristics, accepted positively these systems.

PAGE 66

66 Summary and Findings The analysis of the impact of the two ADAS on this experiment h as shown positive results regarding traffic improvements. In general, the presence and use of the two systems along an arterial followed by a freeway section resulted in higher speeds, smaller space headways and less variability in them during car followin g situations on high speeds The technologies increased the number of lane change maneuvers performed by a driver, and shortened the accepted gaps during lane changing. Besides these traffic impacts, drivers thought these technologies were useful and were satisfied with their performance. These changes in driving behavior indicate that these technologies have the potential to achieve better traffic conditions. Interesting particularities were observed in this dataset. Female drivers were more likely to chan ge their driving behavior than male ones. Younger and older drivers changed their driving behavior, but in different ways. Older drivers improved their headways and performed more lane changes, while younger drivers accepted shorter gaps. Aggressive driver s did not change their driving behavior significantly in any of the case s In contrast the behavior of conservative drivers w as significantly affected by the systems in every case T he presence of the systems seemed to eliminate differences between groups contributing to more uniform and stable traffic conditions.

PAGE 67

67 Table 5 1. Car following data for subject 1 with the ACC Time (sec) Subject's vehicle Lead vehicle Space headway (ft) Speed (mph) Longitudinal Distance (ft) Accel (ft/sec2) Speed (mph) Lo ngitudinal Distance (ft) 582.8 56.0 29169.7 3.6 53.2 29584.9 422.2 583.3 57.2 29211.3 3.5 54.2 29624.3 420.1 583.8 58.3 29253.7 3.3 55.1 29664.4 417.7 584.3 59.4 29297.0 3.1 55.9 29705.1 415.2 584.8 60.5 29341.0 3.0 56.6 29746.4 412.4 585.3 61. 5 29385.8 2.8 57.2 29788.2 409.4 585.8 62.4 29431.2 2.6 57.7 29830.3 406.1 586.3 63.2 29477.3 2.3 58.0 29872.8 402.5 586.8 63.9 29524.0 2.0 57.9 29915.3 398.3 587.3 64.5 29571.2 1.7 57.4 29957.6 393.5 587.8 65.0 29618.7 1.2 56.5 29999.4 387.7 5 88.3 65.3 29666.5 0.6 55.0 30040.2 380.7 588.8 65.4 29714.5 0.0 53.3 30079.9 372.4 589.3 65.3 29762.4 0.6 51.4 30118.2 362.8 589.8 65.0 29810.2 1.2 49.4 30155.1 351.9 590.3 64.5 29857.7 1.7 47.5 30190.5 339.8 590.8 63.8 29904.7 2.2 45.7 3022 4.6 326.9 591.3 63.0 29951.1 2.6 44.1 30257.5 313.3 591.8 62.1 29997.0 2.9 42.6 30289.2 299.2 592.3 61.0 30042.0 3.2 41.1 30319.8 284.8 592.8 59.9 30086.3 3.5 39.6 30349.3 270.0 593.3 58.6 30129.6 3.8 38.1 30377.8 255.1 593.8 57.3 30172.0 4.0 36.8 30405.1 240.1 594.3 55.9 30213.4 4.2 35.5 30431.6 225.1 594.8 54.4 30253.8 4.3 34.4 30457.1 210.3 595.3 53.0 30293.1 4.3 33.5 30482.0 195.9 595.8 51.5 30331.3 4.3 32.9 30506.3 182.0 596.3 50.0 30368.4 4.2 32.5 30530.2 168.8 596.8 48.6 30404.5 4.1 32.4 30554.0 156.5 597.3 47.3 30439.5 3.9 32.5 30577.7 145.2 597.8 46.0 30473.6 3.6 32.9 30601.7 135.1 598.3 44.9 30506.9 3.3 33.5 30626.0 126.2 598.8 43.8 30539.3 2.9 34.3 30650.9 118.6 599.3 42.9 30571.0 2.5 35.4 30676.5 112.5 599.8 42.1 30602.1 2.1 36.6 30702.9 107.8 600.3 41.5 30632.7 1.6 38.0 30730.3 104.6 600.8 41.0 30662.9 1.1 39.5 30758.7 102.8 601.3 40.7 30692.9 0.7 41.1 30788.3 102.5 601.8 40.6 30722.7 0.2 42.7 30819.1 103.4 602.3 40.6 30752.4 0.2 44.3 30851.0 105.6 Average = 267.1 Standard Deviation = 120.7

PAGE 68

68 Table 5 2. Car following aggregate data for each driver, without and with ACC Without ADAS With ADAS Differences Subject ID # of CF* Average Space Headway (ft) Standard D eviation # of CF* Average Space Headway (ft) Standard Deviation Average Space Headway (ft) F test for variances (F test critical=1.31) 1 2 278.48 116.91 4 208.74 96.14 69.74 1.48 2 1 127.88 65.22 2 166.81 75.87 38.93 1.35 3 1 302.93 101.92 3 183 .28 41.61 119.65 6.00 4 1 160.78 84.40 4 187.96 46.96 27.18 3.23 5 3 233.85 112.43 1 181.21 43.88 52.64 6.56 6 1 295.49 96.66 2 196.44 66.23 99.05 2.13 7 2 131.20 43.22 3 179.38 43.91 48.18 1.03 8 1 310.03 154.08 3 209.35 81.32 100.68 3.5 9 9 3 225.69 88.98 2 198.93 59.97 26.76 2.20 10 1 211.15 57.69 4 278.35 133.78 67.20 5.38 11 1 282.47 103.06 1 225.07 144.16 57.40 1.96 12 2 297.62 179.08 2 129.42 49.75 168.20 12.96 13 1 185.24 87.47 3 186.55 66.98 1.31 1.71 14 1 381.67 111.17 2 197.04 58.40 184.64 3.62 15 3 139.70 65.38 1 187.47 49.68 47.77 1.73 16 2 255.37 101.13 2 236.85 96.92 18.52 1.09 17 1 394.18 49.73 2 223.44 84.02 170.74 2.85 18 1 597.22 241.24 3 246.50 156.47 350.72 2.38 19 4 161.37 86.56 3 153. 05 44.83 8.32 3.73 20 2 110.71 64.65 3 168.52 90.92 57.81 1.98 21 2 241.20 147.21 4 208.56 79.95 32.64 3.39 22 1 179.33 91.73 2 156.34 77.86 22.99 1.39 23 3 124.42 65.05 1 248.56 125.76 124.14 3.74 24 1 186.05 70.17 2 202.63 90.92 16.58 1.68 25 4 182.77 183.25 5 176.27 46.60 6.50 15.46 Average = 42.40 *Number of car following situations

PAGE 69

69 Table 5 3. Detailed car following analysis for subjects 19 and 25 Without ACC With ACC Subj ID Car following situation Average space headway (ft) Standard Deviation Car following situation Average space headway (ft) Standard Deviation 19 1 158.54 40.00 1 165.77 42.10 2 114.65 32.07 2 168.75 20.62 3 118.55 41.30 3 136.22 46.62 4 237.35 56.71 Standard Deviation 56.95 17.98 25 1 115.81 34.49 1 116.11 18.71 2 154.92 51.45 2 193.64 38.10 3 105.67 43.96 3 201.27 22.58 4 252.51 44.75 4 215.65 18.62 5 180.35 30.23 Standard Deviation 66.98 38.67 Figure 5 1. Distribution of headway without and with ACC

PAGE 70

70 Figure 5 2. Vehicle trajectory for a car following situation (Subject 14) without ACC Figure 5 3. Vehicle trajectory for a car following situation (Subject 14) with ACC 22000 24000 26000 28000 30000 530 540 550 560 570 580 590 600 Distance (ft) Time (sec) Lead vehicle Subject's vehicle 31000 32000 33000 34000 35000 630 640 650 660 670 Distance (ft) Time (sec) Lead vehicle Subject's vehicle

PAGE 71

71 Table 5 4 Average headway (ft) and standard deviation for gender groups Without With Difference Average space headway (ft) Standard deviation Average space headway (ft) Standard deviation Average space headway (ft) F test (F critical=1.21) Female 245.79 88.91 198.48 79.98 47.31 1.24 Male 236.55 110.52 196. 90 77.07 39.65 2.06 Difference and F test 9.24 1.55 1.58 1.08 Table 5 5 Average headway (ft) and standard deviation for age groups Without With Difference Age Average space headway (ft) Standard deviation Average space headway (ft) Standar d deviation Average space headway (ft) F test (F critical=1.21) 25 30 203.55 84.91 208.46 86.50 4.91 1.04 31 60 284.87 131.17 188.83 71.52 96.04 3.36 Difference and F test 81.32 2.39 19.62 1.46 Table 5 6 Average headway (ft) and standard deviation for driver behavior types Without With Difference Average space headway (ft) Standard deviation Average space headway (ft) Standard deviation Average space headway (ft) F test (F critical=1.21) Aggressive 190.99 106.44 188.56 76.39 2. 43 1.94 Average 182.29 84.14 182.78 63.16 0.49 1.77 Conservative 320.16 115.02 215.46 91.29 104.70 1.59 Aggr Avg 8.70 1.60 5.78 1.46 Avg Conser 137.87 1.87 32.68 2.09 Aggr Conser 129.17 1.17 26.90 1.43

PAGE 72

72 Table 5 7 Number of lane change maneuvers per subject, without and with LCA Without LCA With LCA Difference Subject ID Art Fwy Total Art Fwy Total Art Fwy Total 1 5 3 8 9 2 11 4 1 3 2 10 0 10 12 0 12 2 0 2 3 6 1 7 7 3 10 1 2 3 4 7 3 10 13 1 14 6 2 4 5 10 3 13 7 2 9 3 1 4 6 4 1 5 3 0 3 1 1 2 7 7 2 9 4 4 8 3 2 1 8 6 1 7 8 1 9 2 0 2 9 8 2 10 10 0 10 2 2 0 10 4 1 5 9 0 9 5 1 4 11 4 0 4 3 0 3 1 0 1 12 10 1 11 12 8 20 2 7 9 13 5 0 5 6 3 9 1 3 4 14 6 0 6 6 2 8 0 2 2 15 6 2 8 10 1 11 4 1 3 16 6 2 8 6 0 6 0 2 2 17 6 1 7 4 1 5 2 0 2 18 5 1 6 7 2 9 2 1 3 19 9 5 14 14 4 18 5 1 4 20 6 2 8 8 4 12 2 2 4 21 5 2 7 7 4 11 2 2 4 22 5 0 5 11 1 12 6 1 7 23 11 4 15 14 2 16 3 2 1 24 10 2 12 6 0 6 4 2 6 25 14 4 18 13 4 17 1 0 1 Average 7.0 1.7 8.7 8.4 2.0 10.3 1.36 0.24 1.6 *Art Arterial, Fwy Freeway Table 5 8 Number of lane change maneuvers for gender groups Without With Difference Arterial Freeway Total Arterial Freeway Total Arterial Freeway Total Female 5.9 1.8 7.7 7.8 1.0 8.8 1.9 0.8 1.1 Male 7.6 1.7 9.3 8.7 2.5 11.2 1.1 0.8 1.9 Diff. 1.7 0.1 1.6 0.9 1.5 2.4

PAGE 73

73 Table 5 9 Number of lane change maneuvers by age groups Without With Difference Age Arteri al Freeway Total Arterial Freeway Total Arterial Freeway Total 25 30 6.3 1.5 7.7 7.5 1.6 9.2 1.3 0.2 1.5 31 60 7.6 1.9 9.5 9.0 2.2 11.2 1.4 0.3 1.7 Difference 1.3 0.5 1.8 1.5 0.6 2.0 Table 5 10 Number of lane change maneuvers for d river behavior groups Without With Difference Arterial Freeway Total Arterial Freeway Total Arterial Freeway Total Aggressive 9.4 2.3 11.7 10.9 2.4 13.3 1.4 0.1 1.5 Average 7.3 2.4 9.6 9.1 2.0 11.1 1.9 0.4 1.6 Conservative 5.1 0.8 5.9 6.0 1.6 7.6 0 .9 0.8 1.7 Aggr Avg 2.2 0.1 2.1 1.7 0.4 2.2 Avg Conser 2.2 1.6 3.7 3.1 0.4 3.5 Aggr Conser 4.3 1.5 5.8 4.9 0.8 5.7

PAGE 74

74 Table 5 1 1 Minimum distances (ft) for lane changes performed on the arterial Subject Without With ID BACK FRONT TOTAL BACK FRONT TOTAL 1 2.23 104.81 121.04 22.05 22.85 128.14 2 25.30 32.66 122.03 45.82 41.91 126.84 3 176.91 43.27 266.25 41.20 28.31 127.75 4 32.24 14.20 126.65 35.41 43.13 128.60 5 39.21 31.97 129.56 49.20 23.72 127.73 6 176.39 18.06 271.61 4 6.50 7.87 126.38 7 92.93 20.66 227.03 67.61 43.28 124.89 8 135.23 12.43 161.66 165.71 27.03 216.55 9 27.73 31.69 207.98 41.36 22.27 195.96 10 67.60 42.36 123.96 42.79 48.10 129.12 11 99.28 30.42 143.70 63.95 27.20 126.17 12 31.06 54.38 127.91 3 9.56 33.94 128.60 13 207.29 14.49 235.78 51.53 19.58 120.60 14 46.39 31.05 126.70 137.75 17.66 176.05 15 64.76 17.27 119.99 41.24 20.06 125.67 16 60.63 35.14 223.74 46.51 61.45 121.96 17 373.22 59.36 656.53 80.70 55.21 149.91 18 92.81 36.90 522. 62 161.69 55.90 488.60 19 54.70 51.79 120.49 33.39 29.84 122.37 20 37.38 15.09 121.50 41.14 9.33 119.98 21 63.14 33.04 120.31 61.40 33.65 135.36 22 119.98 42.34 347.42 13.73 11.37 122.29 23 48.10 9.82 131.11 25.23 21.76 125.85 24 14.90 8.47 85. 84 79.47 33.17 126.64 25 38.54 21.75 126.61 38.98 35.10 126.88 Average 85.12 32.54 198.72 58.96 30.95 149.96

PAGE 75

75 Table 5 1 2 Minimu m distances (ft) for lane changes performed on the freeway Subject Without With ID BACK FRONT TOTAL BACK FRONT TOT AL 1 127.97 82.77 224.74 161.79 20.98 196.77 2 3 239.32 629.26 882.58 91.25 40.21 216.97 4 175.42 33.26 222.68 130.98 92.55 237.53 5 142.97 65.42 222.39 180.00 25.96 219.96 6 208.32 622.38 844.70 7 110.19 97.09 221.28 68.04 31.44 202.73 8 205.45 775.99 995.44 200.36 23.85 238.21 9 119.91 78.70 218.68 10 302.66 375.35 692.01 11 12 74.12 146.31 234.43 42.14 18.10 176.15 13 24.34 33.90 120.81 14 190.66 26.16 237.58 15 146.92 38. 16 199.08 50.60 16 143.74 35.62 193.36 17 89.86 135.33 239.19 107.96 116.43 238.39 18 971.11 371.17 41.44 426.61 19 90.27 73.42 217.29 44.24 46.04 187.26 20 174.71 28.34 217.05 108.79 53.86 211.78 21 142.61 63.20 219.81 61.30 31. 85 215.35 22 143.31 16.44 173.75 23 72.61 101.71 235.55 76.45 40.18 212.96 24 165.21 37.21 216.42 25 110.89 27.77 196.43 48.25 76.91 209.98 Average 149.64 220.92 352.27 120.65 43.72 218.99 *Dashes in the table represent no lane chan ges performed or no presence of the specific vehicle in the vicinity. Table 5 1 3 Average of minimum distances (ft) for arterial for gender groups Without With Difference BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL Female 108.8 43.4 245 .8 43.3 28.7 135.4 65.5 14.8 110.4 Male 71.8 26.4 172.2 67.8 32.2 158.1 4.0 5.8 14.1 Difference 37.0 17.0 73.6 24.5 3.5 22.7

PAGE 76

76 Table 5 1 4 Average of minimum distances (ft) for freeway for gender groups Without With Difference BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL Female 132.4 152.3 305.3 114.3 50.1 199.0 2.0 23.9 19.6 Male 159.7 257.8 379.7 122.6 41.3 225.1 44.1 223.9 150.6 Difference 27.3 105.5 74.4 8.3 8.8 26.1 Table 5 1 5 Average of minimum distances (ft) for arterial for age groups Without With Difference Age BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL 25 30 92.3 27.0 184.1 48.6 31.6 126.0 43.7 4.6 58.1 31 60 79.5 36.9 210.2 67.1 30.4 168.8 12.4 6.5 41.5 Diffe rence 12.9 9.8 26.1 18.5 1.2 42.7 Table 5 1 6 Average of minimum distances (ft) for freeway for age groups Without With Difference Age BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL 25 30 165.4 172.2 357.5 77.4 34.9 190.4 66 .9 147.0 177.8 31 60 138.2 253.4 348.5 144.2 49.3 234.6 18.6 182.36 94.8 Difference 27.2 81.2 9.1 66.8 14.4 44.1 Table 5 1 7 Average of minimum distances (ft) for arterial for driver behavior groups Without With Difference BACK F RONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL Aggressive 32.1 35.6 119.2 41.8 29.8 126.9 9.6 5.8 7.7 Average 58.1 30.4 188.0 41.0 30.5 133.0 17.1 0.2 55.1 Conservative 143.8 32.1 262.9 85.3 32.1 179.6 58.5 0.1 83.3 Aggr Avg 26.0 5.2 68.8 0.7 0.7 6.0 Avg Conser 85.7 1.8 74.9 44.3 1.5 46.7 Aggr Conser 111.7 3.5 143.7 43.6 2.2 52.7 Table 5 1 8 Average of minimum distances (ft) for freeway for driver behavior groups Without With Difference BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL Aggressive 116.3 72.3 217.8 82.2 41.4 199.0 14.2 38.0 23.8 Average 136.7 58.8 216.1 112.6 44.4 205.5 32.3 9.5 8.3 Conservative 198.0 510.4 645.6 149.6 44.8 242.0 54.1 464.2 357.0 Aggr Avg 20.5 13.5 1. 7 30.4 3.0 6.5 Avg Conser 61.3 451.5 429.5 37.0 0.5 36.5 Aggr Conser 81.8 438.1 427.8 67.4 3.5 43.0

PAGE 77

77 Table 5 1 9 Average distances (ft) for lane changes performed on the arterial Without With Subj ID BACK FRONT T OTAL BACK FRONT TOTAL 1 241.2 267.2 545.1 199.1 184.4 245.1 2 201.5 239.7 455.2 204.4 166.7 385.1 3 352.9 346.0 708.7 193.3 173.6 376.2 4 174.8 145.8 342.4 132.6 121.7 273.3 5 211.9 246.3 480.9 225.4 209.7 452.9 6 430.2 231.4 728.9 96.9 67.9 147.8 7 235.9 200.7 474.4 140.7 89.2 244.0 8 336.9 282.1 693.9 340.8 216.9 590.7 9 180.4 149.6 350.0 310.9 184.8 527.8 10 223.5 132.2 371.2 219.8 139.6 369.7 11 224.9 200.8 371.9 171.5 128.6 223.2 12 245.8 237.8 488.0 181.0 150.9 347.2 13 423. 4 267.5 699.0 145.9 97.0 256.9 14 232.8 233.7 483.2 282.9 108.3 426.3 15 191.1 117.6 326.4 137.6 142.6 265.9 16 249.0 312.2 575.2 310.0 232.9 556.9 17 350.6 279.0 678.8 287.1 247.9 535.9 18 408.5 391.7 900.6 371.7 332.5 762.0 19 196.8 238.6 449 .5 200.3 152.9 367.2 20 228.3 188.8 431.2 232.5 167.7 424.6 21 144.7 215.0 338.5 270.1 271.3 560.6 22 333.5 308.8 690.4 178.3 169.3 361.6 23 204.8 195.3 374.9 159.7 147.1 320.8 24 130.1 163.5 308.6 226.0 184.5 450.9 25 142.1 117.0 261.9 121.1 116.9 252.0 Average 251.8 228.3 501.1 213.6 168.2 389.0 Table 5 20 Average of distances (ft) for gender groups Without With Difference BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL Female 266.4 233.9 524.0 210.2 167.9 359.0 56.2 6 6.0 165.0 Male 243.6 225.2 488.3 215.5 168.4 405.8 28.1 56.8 82.5 Difference 22.8 8.7 35.7 5.3 0.4 46.8 Table 5 2 1 Average of distances (ft) for age groups Without With Difference Age BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL 25 30 246.7 223.6 476.3 195.7 161.4 362.4 51.0 62.2 113.9 31 60 255.8 232.1 520.7 227.6 173.5 409.8 28.2 58.5 110.8 Difference 9.1 8.5 44.4 31.9 12.1 47.4

PAGE 78

7 8 Table 5 2 2 Average of distances (ft) for driver behavior groups Without With Difference BACK FRONT TOTAL BACK FRONT TOTAL BACK FRONT TOTAL Aggressive 193.8 191.2 394.3 175.6 156.2 323.9 18.2 35.0 70.4 Average 226.3 223.9 474.2 216.3 166.0 401.0 10.0 57.8 73.2 Conservative 312.8 257.9 597.5 238.0 178.4 424.9 74.8 79.6 172.5 Aggr Avg 32.6 32.7 79.9 40.8 9.9 77.2 Avg Conser 86.5 34.1 123.2 21.7 12.3 23.9 Aggr Conser 119.1 66.8 203.2 62.4 22.2 101.1 Table 5 2 3 Average speed (mi/hr) for each subject Subject Without ADAS With ADAS ID Arterial Freeway Total Arterial Freeway Total 1 25.4 52.2 36.7 26.0 52.4 37.3 2 27.0 53.6 38.5 27.9 54.2 39.4 3 25.8 53.7 37.5 27.4 54.8 39.2 4 26.6 53.3 38.1 28.2 55.1 39.9 5 26.5 53.4 38.0 26.8 53.3 38.3 6 24.0 54.1 36.1 24. 6 52.6 36.2 7 25.2 52.9 36.8 25.1 53.3 36.8 8 23.8 53.8 35.8 24.5 52.8 36.2 9 24.3 51.9 35.7 24.9 52.6 36.4 10 24.5 53.6 36.4 26.3 53.7 38.0 11 24.2 53.2 36.0 24.7 53.0 36.4 12 25.3 54.2 37.2 26.8 53.7 38.3 13 25.7 53.3 37.3 27.4 54.2 39.0 1 4 24.8 52.8 36.4 24.9 52.6 36.4 15 26.7 52.9 38.0 27.5 53.5 38.8 16 25.3 52.7 36.8 26.3 52.9 37.7 17 22.4 50.5 32.2 21.6 50.9 31.5 18 22.4 51.1 33.8 21.3 52.1 31.6 19 25.3 53.7 37.1 28.2 53.0 38.1 20 25.8 53.2 37.4 27.4 52.8 37.3 21 25.6 52.3 35.7 26.2 53.9 36.6 22 24.6 52.6 36.2 26.3 53.1 36.6 23 27.3 52.6 38.4 28.9 54.0 38.9 24 28.2 54.3 38.4 26.5 52.5 36.6 25 27.7 53.0 37.7 27.6 54.2 37.9 Average 25.4 53.0 36.7 26.1 53.2 37.2

PAGE 79

79 Table 5 2 4 Average of average speed (mi/hr) for gender groups Without ADAS With ADAS Difference Arterial Freeway Total Arterial Freeway Total Arterial Freeway Total Female 24.7 52.6 36.1 25.6 52.6 36.5 0.9 0.0 0.5 Male 25.8 53.2 37.1 26.4 53.6 37.5 0.7 0.4 0.4 Difference 1.1 0.5 1.0 0.9 0.9 1.0 Table 5 2 5 Average of average speed (mi/hr) for age groups Without ADAS With ADAS Difference Age Arterial Freeway Total Arterial Freeway Total Arterial Freeway Total 25 30 25.7 53.1 37.0 26.6 53.5 37.7 0.9 0.4 0.6 31 60 25.1 52.9 36.5 2 5.8 53.0 36.8 0.7 0.1 0.3 Difference 0.6 0.2 0.6 0.8 0.5 0.9 Table 5 2 6 Average of average speed (mi/hr) for driver behavior groups Without ADAS With ADAS Difference Arterial Freeway Total Arterial Freeway Total Arterial Freeway Total Aggre ssive 26.8 53.3 37.8 27.3 53.5 38.2 0.5 0.2 0.3 Average 25.5 53.0 37.0 26.6 53.3 37.6 1.2 0.3 0.6 Conservative 24.3 52.8 35.7 24.9 53.1 36.1 0.6 0.2 0.4 Aggr Avg 1.3 0.3 0.8 0.7 0.2 0.6 Avg Conser 1.1 0.1 1.3 1.8 0.2 1.5 Aggr Conser 2.5 0.4 2.1 2.4 0.4 2.1 Table 5 2 7 Average user acceptance for all subjects Cronbach's alpha Usefulness Satisfaction 0.89 0.98 1.02 LCA 0.90 1.34 0.65 Combination 0.90 1.17 1.00

PAGE 80

80 CHAPTER 6 CONCLUSIONS This chapter summarizes the research conducted in this thesis and presents the conclusions draw n from the experiments F uture research and recommendations are also offered. Research Summary The thesis evaluated the impacts of two ADAS (Adaptive Cruise Control and Lane Change Assist) in a driving simula tor environment This evaluation was divided in traffic impact of the systems, and user acceptance of them. These systems were implemented in the simulator by adapting algorithms from the literature S cenario s were created in the simulator to provide speci fic situations for using these systems. Once the simulator software was prepared the main objective was accomplished by having drivers driving the scenario first without these systems and secondly using both of them. The observed difference in driver beha vior due to the systems was afterward analyzed by comparing various traffic performance measures U ser acceptance of these systems was also evaluated using a questionnaire. R esults show significant changes in driving behavior that could affect traffic dyna mics positively and also a positive acceptance of these systems by drivers. Research Conclusions This research is an addition to better understand how drivers perceive different types of vehicle technologies. The experiment performed in t his thesis evalua ted two systems integrated in one vehicle and it provides a broad evaluation of ADAS impacts and how they could affect traffic conditions.

PAGE 81

81 The research conclusions based on the traffic impacts of these two systems are presented here. Along the simulated f reeway, participants in general had smaller space headways with the lead vehicle and less headway variability (determined by standard deviation) during car following situations with the ACC harmonizing traffic The division in groups showed that all types of driver s change d their behavior, but older drivers and conservative ones were more affected by this system Changes in driving behavior for these drivers represent a great possibility of traffic improvement The system also contributed to more stable an d homogeneous traffic conditions by eliminating The number of lane change maneuvers was significantly higher on the arterial and as a total for participants in general when using the LCA This patt ern indicates that drivers may have f elt more relaxed to change lanes when the system was available. On the freeway, this pattern was not observed most likely because the ACC was being used at the same time The ACC probably provid ed comfort to drivers since t hey did not have to perform tasks such as accelerating and braking All types of driver s changed their behavior in some aspect except aggressive and young ones These two types of drivers we re more likely to change lane more often than others in normal co nditions (without the LCA) Older drivers male and conservative ones were the most affected by this system increasing the number of lane changes performed Minimum accepted distances with the adjacent vehicle behind and as a total were shorter for partic ipants in general on the arterial when LCA was available. Female drivers and young ones decreased their minimum accepted distances on the arterial

PAGE 82

82 Minimum accepted distances on the freeway were shorter with LCA in all three cases (total gap, distance with adjacent vehicle in front and behind subject vehicle) for participants in general with the LCA Male and conservative drivers w ere the most affected by the system on the freeway Differences between driver behavior types on how they perceived minimum dist ances on the freeway were eliminated when the system was used. Average distances (total gap, distance with vehicle in front and behind the subject vehicle) during lane change maneuvers were shorter for participants in general. Aggressive drivers did not ch ange their behavior regarding accepted gaps while conservative ones were the most affected. T he system eliminated differences between all driver behavior types when LCA was used contributing to more homogeneous traffic Average speed was higher for parti cipants in two of all three cases (arterial and as total). Unexpectedly, average drivers were the only ones that significantly increased their speed with the systems. Differences between female and male groups and among driver types were eliminated when th e systems were available. Analysis on acceptance indicated that drivers were satisfied with the systems and thought of them as useful instruments to support driving tasks in the simulator This indicates that a great percentage of drivers would be willing to use these systems in the future In summary the presence and use of the two ADAS provided, in general, shorter headways and less variability of them under car following, more lane change maneuvers, with shorter minimum and average accepted distances wi th adjacent vehicles, and higher average speeds. In some cases, the systems eliminate significant

PAGE 83

83 differences between groups, providing more uniform and stable traffic conditions. These are all signs of change in driving behavior due to technologies that c ould be used or even adjusted to achieve better traffic conditions by just changing how drivers drive. In this experiment, the systems contributed significantly to improve traffic. Future Research and Recommendations The following recommendations and direc tions for future research are o f fered here. A validation study comparing on road measures with the ones obtained in the simulator should be performed to refine results from this study. An on road experiment with greater sample could quantify differences in traffic performance measures due to these systems. The driving simulator was an appropriate tool to analyze driving behavior changes, but is limited to one vehicle. The traffic impact analyzed for one vehicle in this experiment should be extended to diffe rent market penetrations in a traffic simulator. This would make possible the evaluation of changes in traffic dynamics. Before conducting an experiment as the one on this study, the researcher should analyze in detail the limitations and capabilities of t he driving simulator that will be used. Specific facilities and a dynamic traffic could not be incorporated in this study because of these limitations. Following studies should analyze additional facilities and a more realistic traffic. The sound used for LCA was a probable constraint to the full acceptance of this system. B efore t he actual testing of the system, a human factor analysis should be performed with different indicators.

PAGE 84

84 Safety studies should focus on specific types of drivers to evaluate how th eir driving behavior is affected by these systems. As an example shown in this thesis, aggressive drivers rarely changed their driving behavior due to the systems. Collaboration with other study areas, like the one performed with the Occupational Therapy i n this study, should be encouraged because it results in a more complete understanding of this topic. These systems should be tested with specific driver types that may include older drivers with decreased awareness due to cognitive aging.

PAGE 85

85 APPENDIX A COD IMPLEMENTATION Author: Hariharan Sridharan and Alok Whig This document is intended to be used as a starting point to develop Open Module based Visual Basic modules for the driving simulator STISIM Drive. It also documents the two algorithms (Adaptive Cruise Control and Lane Change Assist) already implemented. The Open Module Programming Guide (provided by STISIM) is the main bible and reference through which all modules may be written (all references to the word guide in this document refer to the Open Module Programming Guide). All programming can only be based out of a computer on which the simulator is installed. This is because there are registered DLLs that come with the simulator which have to be linked with any code you may write. It is essential to read the following topics in the guide before proceeding: Introduction, Open Module Programming and Visual Basic STISIM Drive Variables and Objects Open Module Methods A dinner table reading of the other topics would suffice initially. W hile reading the above topics or once done reading, there are two projects provided with the simulator you should go through. They are located inside \ STISIM \ must copy these folders to a nother folder location on the simulator before proceeding. Play around with the code contained in the copied versions of these modules, link them with the simulator and have fun!

PAGE 86

86 Building your own Open Module Open Module Programming and Vi sual Basic every bit of information to build your own module. For more information, read the Hints and Tips section in this document. In case you are continuing with the Assist Driving module for Cruise Control or ACC, you could cont inue building the Assist_Driving module I have built for the Lane Change/Merge Assist feature. Lane Change/Merge Assist The code for this feature is located in C: \ STISIM \ Open Module \ Assist_Driving To implement this feature, I have modified the OM_Sample _New project. The details of the changes made are: New beep file added to STISIM sounds at C: \ STISIM \ Sound \ beep 4.wav Changed line of code to point to this beep file in the Initialize method Added the following to the Update Method: Dim switchOnSound As B oolean If (Driver.Right = 0) Then If ((switchOnSound = False) And (Events.Vehicles(EIndex(I)).InitialHeading = 0) And (Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) > 8 And (Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 15) Then If ((EDist(I) > 63) And (EDist(I) < 13)) Then switchOnSound = True Open "C: \ \ STISIM \ \ Open Module \ \ LCA.txt" For Append As #17 Print #17, "LEFT Indicator ON at time = & DV.TimeSinceStart & vbNewLine Close #17 End If End If ElseIf (Driver.Left = 0) Then If ((switchOnSound = False) And (Events.Vehicles(EIndex(I)).InitialHeading = 0) And

PAGE 87

87 (Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 8 And (Events.Vehicles(EI ndex(I)).LatPos Vehicle.YLanePos) > 15) Then If ((EDist(I) > 63) And (EDist(I) < 13)) Then switchOnSound = True Open "C: \ \ STISIM \ \ Open Module \ \ LCA.txt" For Append As #17 Print #17, "RIGHT Indicator ON at time = & DV.TimeSinceStart & vbNewLine Close #17 End If End If Else 'do no thing don't beep End If The code above contains distance extents based on which the beep will sound. Adaptive Cruise Control The ACC is divided into two parts. Each version of the algorithm is deployed depending on constrains it satisfies. The fo remost condition required to decide what version of ACC to use (simple or advanced) is to determine whether there is any leading feet. The simple ACC is deployed whe n the closest leading vehicle is more than 525 feet away from driver. This version enables driver to steadily increase its speed with a limit of top speed as set by the driver before starting a simulation. The driver (simulated vehicle) thinks that the roa d is clear of any traffic or obstruction. Common output structure to carry results is of type NODE defined as: Public Type NODE NewSpeed As Double NewAcl As Double End Type Dim NewParams As NODE Given below is the block that executes simple ACC

PAGE 88

88 Private Function Apply_Simple_ACC(MyVehicle As DYNAMICSParams, DV As OMDynamicVariables, ACCFlag As Boolean, MaxSpeed As Double) As NODE Dim Vma acceleration. acceleration and speed. Vmax = MaxSpeed 'Temporary copying for intermediate calculations. CurV = MyVehicle.U If AC CFlag Then 'Compute the acceleration term Acl = (Vmax CurV) / 5 If (Acl > 10) Then Acl = 10 ElseIf (Acl < 20) Then Acl = 20 Else 'Let it remain as it is End If 0.05 sec for 20fps) If CurV Then CurV = CurV + Acl DV.TimeIn c Else CurV = 0 End If If ((Math.Abs(CurV Vmax) / Vmax 100 < 1)) Then CurV = Vmax Acl = 0 End If End If 'P ass on parameters to main loop by initialing the output variable of type structure Params.NewAcl = Acl Params.NewSpeed = CurV

PAGE 89

89 Apply_Simple_ACC = Params End Function Given below is the advanced variant of ACC. Private Function Apply_Adv_ACC(MyVehicle As DYNAMICSParams, DV As OMDynamicVariables, Vp As Double, R As Double, counter As Long, switchOnSound As Boolean, MaxSpeed As Double, Th As Double, ACCSwitch As Boolean) As NODE Static CurV As Double Static Vc As Double Static Rh As Double Static T As Double Static RDot As Double leading vehicle in dri Dim Acl As Double Dim Vmax As Double driver. Static Target_Diff_Percent As Double Stat ic Target_Diff_Speed As Double Dim Cond1 As Boolean Dim Cond2 As Boolean Dim Cond3 As Boolean Dim percent As Double Dim Params As NODE ac celeration of driver. CurV = MyVehicle.U Acl = MyVehicle.UDot 'Dynamics Longitudinal acceleration Vmax = MaxSpeed driver.

PAGE 90

90 I f ((ACCSwitch = True) And CInt(CurV)) Then 'A useful control parameter T = 11 'Step #1 Compute the speed difference between closest leading vehicle and driver. RDot = Vp CurV 'St ep #2 compute Rh ..desired headway distance ... it varies too Rh = Vp Th 'Step #3 Compute Vc commanded velocity for driver. Vc = Vp + (R Rh) / T 'Step #4 Find Acceleration Needed when simulator approaches leading vehicle If ((Math.Abs(CurV Vp) / Vp 100 < 1) And ((Math.Abs(R Rh) / Rh 100 < 1))) Then CurV = Vp Acl = 0 Else 'Test and Apply following. 'Cond 1 Target_Diff_Percent = Math.Abs((R Rh)) / Rh 100 If (Target_Diff_Percent > 1) Then Cond1 = True Else Cond1 = False End If 'Cond 2 Target_Diff_Speed = Math.Abs((Vc CurV)) / Vc 100 If (Target_Diff_Speed > 5) Then Cond2 = True Else Cond2 = False End If 'Cond 3 If ((Vc Vp) > (CurV Vc)) Then Cond3 = True Else Cond3 = False End If If ((Cond1 = True) Or (Cond2 = True) Or (Cond3 = True)) Then 'Acclerration formula to be applied. Acl = (Vc CurV) / 5 Else Acl = Vp CurV

PAGE 91

91 End If End If If (Acl > 4) Then 'If (Acl > 0.25 ) Then Acl = 4 ElseIf (Acl < 20) Then Acl = 20 Else 'let it remain same End If '0.05 is the time step at frame rate 20 /sec check configuration settings C urV = CurV + Acl DV.TimeInc If (CurV > Vmax) Then CurV = Vmax End If End If structure and return control. Params.NewAcl = Acl Params.NewSpeed = CurV Apply_Adv_ACC = Params End Function After the ACC computes acceleration and speed for driver, actual adjustments take place in the Dynamics() function of the STITIM system engine. It is here that dynamic variab les are allowed to be modified. Following shows Dynamics function where modification take place Public Function Dynamics(Dyn As DYNAMICSParams) As Boolean Function for handling all Open Module dynamic updates Parameters: Dyn User defined t ype containing the driver's vehicle dynamic variables Returns: True if everything initialized fine, otherwise false. If false use the ErrorMessage parameter to return a message that the program can display to the user Set the error handling On Error GoTo ErrorOccurred

PAGE 92

92 Dim CurSpeed As String If (Driver.Throttle) Then ThrottleFlg = True Else ThrottleFlg = False End If 'Modify the simulator speed as changed by ACC in Update() If ((NewParams.NewSpeed > 1) And (Dyn.U > 0) And (ACCSwitch = True) And (ThrottleFlg = False)) Then ACC adjusts speed. Dyn.D Th = Driver.Throttle Dyn.U = NewParams.NewSpeed 'Modifying velocity dependent parameters Dyn.AbsU = NewParams.NewSpeed Dyn.UDot = NewParams.NewAcl Else NewParams.NewSpeed = Dyn.U NewParams.NewAcl = Dyn.Ax End If If StaticVars.DisplaySystem = "CenterDisplay" Then CurSpeed = CStr(CInt(Dyn.U 0.681)) With FormSpeed '.Height = Scr een.Height / 10 '.Width = Screen.Width / 10 .lblSpeed.FontSize = 25 .lblSpeed = CurSpeed .BorderStyle = 0 End With End If Setup the return from function Dynamics = True Exit Function Handle any errors ErrorOccurred: Error Message = ProcessError("Dynamics") Dynamics = False End Function The other important function is the Update() where logic of determining traffic flow and also to filter traffic to determine vehicles that should be considered for ACC calculations.

PAGE 93

93 Only the covered here. Public Function Update(DV As OMDynamicVariables, Vehicle As DYNAMICSParams, Events As SimEvents, NumEvents As Integer, EDist() As Single, EDes() As Integer, EI ndex() As Integer, SimSounds() As SoundEffects) As Boolean Function for handling all Open Module action during the actual simulation loop Parameters: DV User defined type containing the simulation parameters that are changing at each time step Vehicle User defined type containing the driver's vehicle dynamic variables Events UDT that contains the parameter settings for each supported event NumEvents Number of events that are in the current display l ist EDist() Distance from the driver to the event EDes() Event designator for each active event EIndex() Event index for each event in the display list. This value is the index into the Events UDT so that you can get the parameters for each individual event in the display list SimSounds() Contains the sound buffer references to the simulator sound effects Returns: True if everything initialized fine, otherwise false. If false use the ErrorMes sage parameter to return a message that the program can display to the user 'There are three kinds of traffic based on direction of their flow. lowing in same direction to driver. Case 15 checking if turn indicator is switched on If switchOnSound = False And (Events.Vehicles(EIndex(I)).InitialHeading = 0) And (Math.Abs(Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 13) Then If (EDist(I) > 63 And EDist(I) < 13) Then 'Or

PAGE 94

94 (Events.Vehicles(EIndex(I)).Dist < 63 And Events.Vehicles(EIndex(I)).Dist > 0) Then If Driver.Right = 0 Or Driver.Left = 0 Then switchOnSound = True change. Open "C: \ \ STISIM \ \ Open Module \ \ LCA.txt" For Append As #17 Print #17, "Indicator ON at time = & DV.TimeSinceStart & vbNewLine Close #17 End If End If End If 'Task is to identify vehicles in same direction and same lane. 'Assume: One direction traffic that is no approaching vehicle.[No event A] 'Find vehicles in same lane as simulator. 'Find vehicles in same lane that is being followed closest. If counter = 1 Then vehDis(Events.Vehicles(EIndex(I)).VehNum) = Events.Vehicles(EIndex(I)).Dist End If 'To determine traffic direction If (Events.Vehicles(EIndex(I)).CosTerm = 1) Then If (Math.Abs(Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 7) Then vehDis (Events.Vehicles(EIndex(I)).VehNum) = EDist(I) 'Find vehicles that are being followed and that are following sim If (vehDis(Events.Vehicl es(EIndex(I)).VehNum) < 0) Then 'Last task is to find closest vehicle among leading vehicles and then apply ACC 'maintain si mulator speed Else

PAGE 95

95 'replaced Range by EDist. Value 15 is subtracted to compensate for ignored. If (clstLeadVehRng > EDist(I) 15) Then clstLeadVehRng = EDist(I) 15 'Activate ACC if range between simulator and the leading vehicle is < 525 If (clstLeadVehRng < 525) Then 'Call wrapper function that internally calls ACC after reading from 'ACCSwitchFile.txt and further activating ACC() dep ending on switch value. Vp = Events.Vehicles(EIndex(I)).Speed R = EDist(I) 15 'apply the advanced case of ACC. Sp eed and acceleration adjusted. NewParams = Apply_Adv_ACC(Vehicle, DV, Vp, R, counter, switchOnSound, MaxSpeed, Th, ACCSwitch) Else 'apply the simple case New Params = Apply_Simple_ACC(Vehicle, DV, ACCSwitch, MaxSpeed) End If End If End If Else End If Else driver..ignore End If End of Function

PAGE 96

96 Lastly, the head way time (variable Th in ACC) and the maximum allowable speed for driver is supplied through an ext ernal file read inside Initialize() function system provided. Below is an extract from the function where file is read. ACCSwitch = False Open "C: \ \ STISIM \ \ Open Module \ \ ACC_Param.OM" For Input As #1 Line Input #1, snextline snextline = snextline & vbCrLf MaxSpeed = snextline Line Input #1, snextline snextline = snextline & vbCrLf Th = snextline Close #1 All Hints and Tips 1. To get started, it is highly recommended to continue building from one of the two similar to the samples provided in the Update method of these projects. 2. Make sure you use IntelliSe nse in VB to find out what objects or functions you can currently use on the STISIM object you are working with (hit Ctrl reference sections in Appendix A and B (towards the end) are the most useful sections that will tell you what all th ose letters and functions will do. stop shop for any questions you may have about the STISIM objects/methods. It is located at: http://forums.systemstech.com/viewforum.php?f=2 You will need to login to be able to post on the forum. To apply for your own login, register at the forum and then email Ted asking him to activate your credentials. You can expect responses to your queries within 1 after 3 working days, consider emailing Ted only then. 4. To link your Open Module DLL with the simulator, TODO: fill this up at the simulator 5. Every single time you edit the code, you will need to do the following to link it with the simulator. b) Copy it to the same folder locations in the Left and Right systems too as it is currently in the center system. (For Ex: C: \ STISIM \ O pen Module \ My Module on all 3 systems Center, Left and Right). c) Register the DLL on all 3 systems as outlined in the guide.

PAGE 97

97 Code that generated the DLL Option Explicit Public ACCSwitch As Boolean Public MaxSpeed As Double Public Th As Double Public T ype NODE NewSpeed As Double NewAcl As Double End Type Dim NewParams As NODE Public ThrottleFlg As Boolean 'added july 16 Dim Tools As New TJRWinToolsCls Dim Graphics As New TJR3DGraphics Dim Terrain As New STI_3D_Terrain Dim Ds As DirectSound Pub lic ErrorMessage As String Public LogFileHandle As Long Public NewForm As Form Public TextMessage As String Public WillHandleCrash As Long Type DriverControlInputs Steer As Single Steering angle count from the controller card Throttle As Single Throttle control count from the controller card brake As Single Brake control count from the controller card Gear As Integer Current transmission gear Horn As Int eger Current state of the horn button, 0 Activated Left As Integer Current state of the left turn signal, 0 Activated Right As Integer Current state of the right turn signal, 0 Activated End Type Dim Driver As DriverControlInputs Type SoundFiles Active As Boolean True if sound is currently available for playing Buffer As DirectSoundBuffer DirectSoundBuffer object that holds the WAV file to be played FileName As String File name of the WAV file that will be played Running As Boolean True if the current sound effect is being played End Type Dim Sounds() As SoundFiles Type Vehicle SixDof As SixDOF Position Inertial orientation of the vehicle

PAGE 98

98 Lat As Single Lane postion with respect to the roadway centerline Lon As Single Distance into the run (from the beginning) Speed As Single Vehicle speed Acceleration As Single Vehicle acceleration Index(1) As Long 0 Lights off, 1 Lights on VisFlag(1) As Long 0 Lights off, 1 Lights on InitialH eading As Single Vehicle initial heading when it is activated End Type Dim V() As Vehicle Dim Bool As Boolean Temporary boolean variable for use anytime a throw away boolean is needed Dim DataFileName As String Name of the open module data file that will hold data during the run Dim DataFileNum As Integer File number for the data file that will hold data during the run Dim DynVars As OMDynamicVariables UDT containing STISIM Drive varia bles that change as the run progresses Dim ImagePointers() As Long Array containing the handles of the screen images that will be used Dim NumImages As Integer Number of screen images that will be used Dim NumSounds As Intege r Number of sound effect files that will be used Dim NumVehicles As Integer Number of vehicles that will be displayed Dim ScreenObjects() As SixDOFPosition Array of screenobjects that will displayed Dim St As String Temporary string variable for use anytime a throw away string is needed Dim StaticVars As OMStaticVariables UDT containing STISIM Drive variables that are fixed by the simulator Public Function AddNew(OMVars As OMParameters) As B oolean On Error GoTo ErrorOccurred AddNew = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("AddNew") AddNew = False End Function Public Function ControlInputs(Dyn As DYNAMICSParams, Steering As Single, Throttle As Single, brake As Single, Gear As Integer, DInput As Integer) As Boolean On Error GoTo ErrorOccurred Driver.brake = brake Driver.Steer = Steering Driver.Throttle = Throttle Driver.Gear = Gear

PAGE 99

99 Driver.Horn = DInput And 1 Driver.Left = DInput And 2 Driver.Right = DInput And 4 ControlI nputs = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("ControlInputs") ControlInputs = False End Function Public Function Dynamics(Dyn As DYNAMICSParams) As Boolean On Error GoTo ErrorOccurred Dim CurSpeed As String If (Driver.Throttle ) Then ThrottleFlg = True Else ThrottleFlg = False End If If ((NewParams.NewSpeed > 1) And (Dyn.U > 0) And (ACCSwitch = True) And (ThrottleFlg = False)) Then Dyn.DTh = Driver.Throttle Dyn.U = NewParams.NewSpe ed Dyn.AbsU = NewParams.NewSpeed Dyn.UDot = NewParams.NewAcl Else NewParams.NewSpeed = Dyn.U NewParams.NewAcl = Dyn.Ax End If If StaticVars.DisplaySystem = "CenterDisplay" Then CurSpeed = CStr(CIn t(Dyn.U 0.681)) 'convert ft/sec to mph With FormSpeed .lblSpeed.FontSize = 40 .lblSpeed = CurSpeed .BorderStyle = 0 End With End If Dynamics = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("Dynamics") Dynamics = False End Function Public Function HandleCrash(Override As Integer) As Boolean On Error GoTo ErrorOccurred Override = 0 HandleCrash = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("HandleCrash")

PAGE 100

100 HandleCrash = False End Function Dim FileN um As Integer Dim I As Integer Dim SoundFileNames(1) As String Dim snextline As String 'added july 20 On Error GoTo ErrorOccurred StaticVars = SV ACCSwitch = False Open "C: \ \ STISIM \ \ Open Module \ \ ACC_Param.OM" For Input As #1 Line Input #1, snextline snextline = snextline & vbCrLf MaxSpeed = snextline Line Input #1, snextline snextline = snextline & vbCrLf Th = snextline Close #1 Graphics.Renderer = StaticVars.Renderer SoundFileNames(0) = "C: \ STISIM \ Sound \ beep 4.wav" SoundFileNa mes(1) = "C: \ STISIM \ Sound \ Horn.Wav" If StaticVars.DisplaySystem = "CenterDisplay" Then NumSounds = 2 If NumSounds Then If StaticVars.SoundOn = True Then ReDim Sounds(NumSounds 1) For I = 0 To NumSounds 1 Sounds(I).FileName = SoundFileNames(I) If Tools.FileExist(Sounds(I).FileName) = False Then ErrorMessage = "The warning sound file & Sounds(I).FileName & could not be found! The simulation run will be aborte!" Initialize = False Exit Function End If Next For I = 0 To NumSounds 1 Sounds(I).Active = Tools.CreateDSBFromWaveFile(Ds, Sounds(I).FileName, Sounds(I).Buffer ) Next End If End If End If With CommForm.MSComm1 If .PortOpen = False Then .CommPort = 1 .Settings = "9600,N,8,1" .RThreshold = 0 .InBufferSize = 128

PAGE 101

101 .InputLen = 0 .InputMode = co mInputModeText .NullDiscard = False .OutBufferSize = 128 .ParityReplace = "" .RThreshold = 0 .PortOpen = True End If End With With FormSpeed .Height = 1400 .Width = 1800 .Visible = True .Top = Screen.Heig ht 0.4 .Left = 1.25 Screen.Width .BorderStyle = 0 .Caption = "Your Speed" .lblSpeed.Alignment = 0 .lblSpeed.FontName = "Verdana" .lblSpeed.Top = 0 End With Tools.WindowOnTop (FormSpeed.hWnd) ACCSwitch = False Initialize = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("Initialize") Initialize = False End Function Public Function InitializeTerrain(RCL() As RoadCenterLine, NumRoadSegs As Long, Road() As EventROADType, NumRoad As Long, Vc() As EventVCType, NumVC As Long) As Boole an SDL scenario file) On Error GoTo ErrorOccurred Call Terrain.AssignRCLVariables(RCL(), NumRoadSegs) Call Terrain.AssignRoadVariables(Road(), NumRoad) Call Terrain.AssignVCVariables(Vc(), NumVC) Terrain.NumRoadSeg = NumRoadSegs InitializeTerrain = True Ex it Function ErrorOccurred: ErrorMessage = ProcessError("initializeTerrain") InitializeTerrain = False End Function Public Function PostRun(Comments As String, DriverName As String, RunNumber As On Error GoTo ErrorOccurred Set Ds = Nothing Set Tools = Noth ing PostRun = True

PAGE 102

102 Exit Function ErrorOccurred: ErrorMessage = ProcessError("PostRun") PostRun = False End Function Public Function Shutdown(RunCompleted As Integer) As Boolean Dim I As Integer On Error GoTo ErrorOccurred If StaticVars.DisplaySystem = "Cen terDisplay" Then If NumSounds Then For I = 0 To NumSounds 1 If Sounds(I).Active = True Then Sounds(I).Buffer.Stop Set Sounds(I).Buffer = Nothing End If Next End If End If Set Graphics = Nothing Set Terrain = Nothing Shutdown = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("Shutdown") Shutdown = False End Function Public Function StartUp(Config As GAINSParams, BackForm As Object, SV As OMStaticVariables, UseNew As Boolean, DsIn As DirectSound) As Boolean On Error GoTo ErrorOccurred If SV.SoundOn = True Then Set Ds = DsIn DirectSoundCreate ByVal 0&, Ds, Nothing Ds.SetCooperativeLevel SV.ActiveHandle, DSSCL_NORMAL End If UseNew = False StartUp = True Ex it Function ErrorOccurred: ErrorMessage = ProcessError("StartUp") StartUp = False End Function Public Function Update(DV As OMDynamicVariables, Vehicle As DYNAMICSParams, Events As SimEvents, NumEvents As Integer, EDist() As Single, EDes() As Integer, EIndex() As Integer, SimSounds() As SoundEffects) As Boolean Dim ElapsedTime As Single Dim FirstLight As Boolean Dim I As Integer Dim J As Integer Dim LightIndex As Integer

PAGE 103

103 Dim switchOnSound As Boolean Dim snextline As String 'Simulates hardware switch tha t activates ACC(). Dim CurSpeed As Long 'Will be removed later. For testing only whether ctrl goes to ACC Static counter As Long 'Temporary counter used to change ACC flag after a limit to simulate Dim K As Double 'for testing Static FLG As Integer Dim DstTravelled As Long Static clstLeadVehRng As Double 'Closest leading Vehicle Distance to Simulator Static vehDis(250) As Double Static Vp As Double Static R As Double Dim Err As Double Static Range(250) As Double clstLeadVehRng = 100000 'Init with a la rge number On Error GoTo ErrorOccurred If StaticVars.DisplaySystem = "CenterDisplay" Then switchOnSound = False counter = counter + 1 July 22, 2010 for printing ACC state Open "C: \ \ STISIM \ \ Open Module \ \ ACCState.txt" For Append As #29 If ((DV.KeyCommand = 71) Or (DV.KeyCommand = 103)) Then ACCSwitch = True switchOnSound = True Print #29, "ACC is TRUE at time =" & DV.TimeSinceStart & vbNewLine End If If ((A CCSwitch = True) And Driver.brake) Then ACCSwitch = False Print #29, "ACC is FALSE at time = & DV.TimeSinceStart & vbNewLine & vbNewLine End If Close #29 For I = 1 To NumEvents Select C ase EDes(I) Case 15 If (Driver.Right = 0) Then If ((switchOnSound = False) And (Events.Vehicles(EIndex(I)).InitialHeading = 0) And (Events.Vehicles(EIndex(I)).LatPos Vehicle.Y LanePos) > 8 And (Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 15) Then If ((EDist(I) > 63) And (EDist(I) < 13)) Then switchOnSound = True O pen "C: \ \ STISIM \ \ Open Module \ \ LCA.txt" For Append As #17 Print #17, "LEFT Indicator ON at time = & DV.TimeSinceStart & vbNewLine Close #17

PAGE 104

104 End If End I f ElseIf (Driver.Left = 0) Then If ((switchOnSound = False) And (Events.Vehicles(EIndex(I)).InitialHeading = 0) And (Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 8 And (Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) > 15) Then If ((EDist(I) > 63) And (EDist(I) < 13)) Then switchOnSound = True Open "C: \ \ STISIM \ \ Open Module \ \ LCA.txt" For Append As #17 Print #17, "RIGHT Indicator ON at time = & DV.TimeSinceStart & vbNewLine Close #17 End If End If Else End If If counter = 1 Then vehDis(Events.Vehicles(EIndex(I)).VehNum) = Events.Vehicles(EIndex(I)).Dist End If If (Events.Vehicles(EIndex(I)).CosTerm = 1) Th en If (Math.Abs(Events.Vehicles(EIndex(I)).LatPos Vehicle.YLanePos) < 7) Then vehDis(Events.Vehicles(EIndex(I)).VehNum) = EDist(I) If (vehDis(Events.Vehicles(EIndex(I )).VehNum) < 0) Then Else If (clstLeadVehRng > EDist(I) 15) Then clstLeadVehRng = EDist(I) 15 'Discussed on July 13 If (clstLeadVehRng < 525) Then Vp = Events.Vehicles(EIndex(I)).Speed R = EDist(I) 15 NewParams = A pply_Adv_ACC(Vehicle, DV, Vp, R, counter, switchOnSound, MaxSpeed, Th, ACCSwitch) Else NewParams = Apply_Simple_ACC(Vehicle, DV, ACCSwitch, MaxSpeed) End If End If End If Else

PAGE 105

105 End If Else 'Implies the vehicle is of type 'A' or travell ing in opposite direction End If Case 16 If EDist(I) < 100 Then End If Case 17 With Events.Sl(EIndex(I)) LightIndex = EIndex(I) End With Case 21 With Events.Ped(EIndex(I)) If Events.Sl(LightIndex).LightState(1) = 2 Then If Int(.Direction) = 1 Or Int(.Direction) = 4 Then .W alking = 1 .Velocity = (DV.RoadWidthL + DV.RoadWidthR + 4) / (Events.Sl(LightIndex).RedOnTime 3) End If If .Walking Then If .LatPos < (DV.RoadWidthL + 4) Then .Velocity = 0 .Direction = 0 .Walking = 0 End If Else .Dist = 0 End If ElseIf Events.Sl(LightIndex).LightState(1) = 0 Then If Events.Sl(LightIndex).FreezeOnGreen = True Then If Int(.Direction) = 0 Or Int(.Direction) = 3 Then .Walking = 1 .Velocity = 3.5 .Dist = .Dist + .Velocity DV.TimeInc End If If Abs(.Dist) > (DV.RoadWidthL + D V.RoadWidthR + 7) Then .Walking = 0 .Velocity = 0 .Image = 0 .Direction = 4.7124 End If End If End If End With End Select Next Call AuditoryWarning(switchOnSound, 0)

PAGE 106

106 End If With CommForm.MSComm1 If .PortOpen = True Then .Output = "Test" DoEvents End If End With 'De allocate memory from array variables -Added July 27,2010 Erase vehDis, Range DynVars = DV Update = True Exit Function ErrorOccurred: ErrorMessage = ProcessError("Update") Update = False End Function Private Sub AuditoryWarning(Setting As Bo olean, Index As Integer) Dim DSBStatus As Long If Setting = True Then If Sounds(Index).Running = False Then If Sounds(Index).Active = True Then Sounds(Index).Buffer.Play 0, 0, 1 Sounds(Index).Running = True End I f End If ElseIf Sounds(Index).Running = True Then Call Sounds(Index).Buffer.GetStatus(DSBStatus) If DSBStatus <> DSBSTATUS_PLAYING Then Sounds(Index).Running = False Sounds(Index).Buffer.Stop Sounds(Index).Buffer.SetCurr entPosition 0 End If End If End Sub Private Function ProcessError(ModuleName As String) As String St = "Simulation run aborted! An error has occurred in Open Module & ModuleName & ":" & vbCrLf & vbCrLf St = St & "Error number:" & vbTab & Trim(Str(Err .Number)) & vbCrLf St = St & "Description:" & vbTab & Err.Description & vbCrLf St = St & "Error source:" & vbTab & Err.Source & vbCrLf ProcessError = St & "Last DLL Error:" & vbTab & Err.LastDllError & vbCrLf Bool = Tools.WriteToTJRFile(StaticVars.LogFileH andle, ProcessError) End Function Private Function Apply_Simple_ACC(MyVehicle As DYNAMICSParams, DV As OMDynamicVariables, ACCFlag As Boolean, MaxSpeed As Double) As NODE Static CurV As Double 'current speed of driver Dim Vmax As Double 'm aximum speed allowed Dim Acl As Double 'computed acceleration

PAGE 107

107 Dim Params As NODE 'A structure carrying final computed speed and accelration of driver Vmax = MaxSpeed CurV = MyVehicle.U Acl = (Vmax CurV) / 5 If (Acl > 10) Then Acl = 10 ElseIf (Acl < 20) Then Acl = 20 Else End If If CurV Then CurV = CurV + Acl DV.TimeInc Else CurV = 0 End If If ((Math.Abs(CurV Vmax) / Vmax 100 < 1)) Then CurV = Vmax Acl = 0 End If End If Params.NewAcl = Acl Params.NewSpee d = CurV Apply_Simple_ACC = Params End Function Private Function Apply_Adv_ACC(MyVehicle As DYNAMICSParams, DV As OMDynamicVariables, Vp As Double, R As Double, counter As Long, switchOnSound As Boolean, MaxSpeed As Double, Th As Double, ACCSwitch As Boolean) As NODE Static CurV As Double 'current speed of driver Static Vc As Double 'commanded speed Static Rh As Double 'head way distance Static T As Double control paramter Static RDot As Double 'rate of change of speed difference between driver and closest leading vehicle Dim Acl As Double 'final acceleration of driver Dim Vmax As Double 'max allowed speed Static Target_Diff_ Percent As Double 'used as a threshold Static Target_Diff_Speed As Double 'used as a threshold Dim Cond1 As Boolean Dim Cond2 As Boolean Dim Cond3 As Boolean Dim percent As Double 'temp variable Dim Params As NODE 'output parameter carrying final speed and acceleration of driver CurV = MyVehicle.U Acl = MyVehicle.UDot 'Dynamics Longitudinal acceleration

PAGE 108

108 Vmax = MaxSpeed If ((ACCSwitch = True) And CInt(CurV)) Then T = 11 'average RDot = Vp CurV Rh = Vp Th Vc = Vp + (R Rh) / T If ((Math.Abs(CurV Vp) / Vp 100 < 1) And ((Math.Abs(R Rh) / Rh 100 < 1))) Then CurV = Vp Acl = 0 Else Target_Diff_Percent = Math.Abs((R Rh)) / Rh 100 If (Target_Diff_Percent > 1) Then Cond1 = True Else Cond1 = False End If Targe t_Diff_Speed = Math.Abs((Vc CurV)) / Vc 100 If (Target_Diff_Speed > 5) Then Cond2 = True Else Cond2 = False End If If ((Vc Vp) > (CurV Vc)) Then Cond3 = True Else Cond3 = False End If If ((Cond1 = True) Or (Cond2 = True) Or (Cond3 = True)) Then Acl = (Vc CurV) / 5 Else Acl = Vp CurV End If End If If (Acl > 10) Then 'If (Acl > 0.25) Then Acl = 10 ElseIf (Acl < 20) Then Acl = 20 Else End If CurV = CurV + Acl D V.TimeInc If (CurV > Vmax) Then CurV = Vmax End If End If Params.NewAcl = Acl Params.NewSpeed = CurV Apply_Adv_ACC = Params End Function

PAGE 109

109 APPENDIX B SCENARIO FILES Second Scenario 1, output 0, BSAV, 1, 0 .495, SPEED, 1, 23, 2, 6, 7, 36, 19 1, Roadway arterial 0, ROAD, 12, 6, 3, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 800, ROAD, 12, 6, 3, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 2400, ROAD, 12, 6, 3, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 3600, ROAD, 12, 6, 3, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 4400, ROAD, 12, 6, 3, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 6200, ROAD, 12, 6, 3, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 13800, ROAD, 12, 4, 2, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5 17300, ROAD, 12, 4, 2, 2, 0, 10, 10, 0.333, 0.333, 400, 0, 0, 0, 5, 0, 5,0,0,0,0,6 1 curves 16500,C,0,0,500,0, .00025 19000,C,0,0,800,0, .0005 26300, C,0,0,800,0, .0003 35800, C,0,0,800,0, .0005 1, interse ction 4800,I,0,0,3,0,0,2 7800,I,0,0,2,0,0,1 12800,I,0,0,2,0,0,1 15800,I,0,0,3,0,0,2 1, Signals 2400, SL, 2400,15,3,15,0,0,2,1,0 5400, SL, 2400,15,3,15,0,0,2,1,0 10400, SL, 2400,15,3,15,0,0,2,1,0 13400, SL, 2400,15,3,15,0,0,2,1,0 1, Speed limit Signs 0, SIGN, 100, 200, C: \ STISIM \ Data \ Signs \ Sp40Mph.3ds 2200, SIGN, 100, 1000, C: \ STISIM \ Data \ Signs \ Sp40Mph.3ds 4900, SIGN, 100, 400, C: \ STISIM \ Data \ Signs \ Sp40Mph.3ds 7900, SIGN, 100, 400, C: \ STISIM \ Data \ Signs \ Sp40Mph.3ds 10800, SIGN, 100, 1000, C: \ STISIM \ Data \ S igns \ Sp40Mph.3ds 12900, SIGN, 100, 300, C: \ STISIM \ Data \ Signs \ Sp40Mph.3ds 12900,SIGN, 100, 600, C: \ STISIM \ Data \ Signs \ NARROWS.3ds 15900, SIGN, 100, 200, C: \ STISIM \ Data \ Signs \ SP65MPH.3DS 17200, SIGN, 100, 1000, C: \ STISIM \ Data \ Signs \ SP65MPH.3DS 18200, SIGN, 4, 600, 0, 0

PAGE 110

110 23200, SIGN, 100, 1000, C: \ STISIM \ Data \ Signs \ SP65MPH.3DS 25500, SIGN, 4, 600, 0, 0 29200, SIGN, 100, 1000, C: \ STISIM \ Data \ Signs \ SP65MPH.3DS 32000, SIGN, 100, 1000, C: \ STISIM \ Data \ Signs \ SP65MPH.3DS 35000, SIGN, 4, 600, 0, 0 36500, SIGN, 100, 1000 C: \ STISIM \ Data \ Signs \ Sp65Mph.3ds 1 sounds 3500, PR, C: \ STISIM \ Data \ FHWA \ Sounds \ IntersectionLeft.wav,0,10 6500, PR, C: \ STISIM \ Data \ FHWA \ Sounds \ IntersectionRight.wav,0,10 11500, PR, C: \ STISIM \ Data \ FHWA \ Sounds \ IntersectionRight.wav,0,10 14500, PR, C: \ STIS IM \ Data \ FHWA \ Sounds \ IntersectionLeft.wav,0,10 16500, PR, C: \ STISIM \ Data \ FHWA \ Sounds \ accsound.wav,0,10 1,vehicles on the same side 0, V,40,60,6,1, *1~4 0, V,45,150,18,1, *18~27 0, V,45,100,30,1, *31~35 0, V,50,200,30,1, *31~35 0, V,49,260,18,1, *1~4 0, V,50,380,30,1, *31~35 0, V,48,300,6,1, *1~4 0, V,49,450,6,1, *18~27 0, V,52,490,6,1, *18~27 0, V,52,390,18,1, *31~35 0, V,52,460,30,1, *1~4 0, V,53,480,18,1, *1~4 0, V,55,550,6,1, *1~4 0, V,57,530,18,1, *31~35 0, V,55,620,30,1, *18~27 0, V,55,600,6,1, 1~4 0, V,55,700,30,1, *1~4 0, V,61,730,6,1, *18~27 0, V,62,600,18,1, *1~4 0, V,62,750,30,1, *31~35 0, V,63, 500,6,1, *18~27 0, V,/2, 450,18,1, *1~4 0, V,64, 350,30,1, *31~35 0, V,61, 650,6,1, *18~27 0, V,65, 700,18,1, *31~35 0, V,62, 750,30,1, *18~27 0, V, 63, 800,6,1, *1~4 0, V,70, 850,18,1, *31~35 0, V,60, 820,30,1, *1~4 4850, v, 44, 150, 6, 1, *18~27 4850, v, 42, 430, 18, 1, *31~35

PAGE 111

111 4850, v, 44, 250, 30, 1, *1~4, 1, 0, *10, 2 4850, v, 44, 500, 6, 1 *18~27 4850, v, 46, 550, 18, 1, *18~27 4850, v, 48, 400 30, 1, *31~35 4850, v, 50, 600, 6, 1, *18~27 4850, v, 51, 650, 18, 1, *18~27 4850, v, 51, 800, 30, 1, *1~4 4850, v, 53, 630, 6, 1, *18~27 4850, v, 53, 680, 18, 1, *18~27 4850, v, 54, 890, 30, 1, *1~4 4850, v, 44, 200, 18, 1, *31~35 4850, v, 46, 200, 30 1, *18~27 5400, V, 63, 500,6,1, *18~27 5400, V, /2, 450,18,1, *1~4 5400, V, 64, 350,30,1, *31~35 5400, V, 61, 450,6,1, *18~27 5400, V, 65, 300,18,1, *31~35 5400, V, 52, 450,30,1, *18~27 5400, V, 63, 550,6,1, *1~4 7850, v, 44, 200, 6, 1, *18~27 785 0, v, 46, 320, 18, 1, *18~27 7850, v, 44, 330, 30, 1, *18~27 7850, v, 49, 350, 6, 1, *18~27 7850, v, 50, 390, 18, 1, *18~27 7850, v, 50, 500, 30, 1, *18~27 7850, v, 52, 450, 6, 1, *18~27 7850, v, 52, 500, 18, 1, *18~27 7850, v, 53, 700, 30, 1, *18~27 7850, v, 55, 620, 6, 1, *18~27 7850, v, 55, 580, 18, 1, *18~27 7850, v, 57, 820, 30, 1, *18~27 7850, v, 58, 700, 6, 1, *18~27 7850, v, 59, 710, 18, 1, *18~27 7850, v, 60, 850, 30, 1, *1~4 7850, v, 45, 850, 18, 1, *18~27 7850, v, 48, 970, 30, 1, *1~4 7850, v, 65 100, 30, 1, *1~4 7850, v, 48, 200, 18, 1, *18~27 7850, v, 60, 250, 18, 1, *18~27 7850, v, 60, 250, 30, 1, *18~27 7850, v, 65, 350, 6, 1, *18~27 7850, v, 65, 350, 30, 1, *18~27 8400, V, 63, 500,6,1, *18~27 8400, V, /2, 450,18,1, *1~4 8400, V, 64, 350,30,1, *31~35 8400, V, 61, 450,6,1, *18~27

PAGE 112

112 8400, V, 65, 300,18,1, *31~35 8400, V, 52, 450,30,1, *18~27 8400, V, 63, 550,6,1, *1~4 12830, V,73, 800,18,1, *18~27 12830, V,70, 600,6,1, *1~4 12830, V,51,500,6,1, *31~35 12830, V,57,550,18,1, *1~4 12830, V,60,600,6,1, *1~4 12830, V,61,650,18,1, *31~35 12830, V,55,700,6,1, *18~27 1 freeway vehicles 15850, V,95,350,9,1, *1~4 15850, V,103,500,21,1, *18~27 15850, v, 91, 800, 9, 1, *1~4 16000, V,94,850,9,1, *31~35 16450, v, 95, 800, 21,1, *18~27 16950, v, 97 800, 9,1, *31~35 17200, V,88,800,9,1, *1~4 17350, v, 98, 800, 21,1,*31~35 18050, v,97, 800, 21,1, *18~27 18650, v, 92, 800, 21,1,*31~35 19350, v, 91, 800, 21,1,*1~4 19750, V,103,850,21,1, *31~35 19900, V,101,800,9,1, *18~27 20200, V,95,800,9,1, *31~35 2 0350, v, 95, 800, 21,1,*1~4 21200, V,97,800,9,1, *31~35 24800, V,98,800,9,1, *18~27 25200, V,92,900,9,1, *31~35 26850, V,94,850,21,1, *1~4 26900, V,95,900,21,1, *31~35 27800, V,89,800,9,1, *18~27 27900, V,91,900,21,1, *31~35 28500, V,95,800,21,1,*1~4 292 00, V,97,900,9,1, *31~35 29850, V,98,850,21,1, *18~27 30500, V,89,950,9,1, *31~35 30850, V,100,850,21,1, *1~4 31350, v, 95, 900, 21,1,*31~35 31500, V,92,850,9,1, *31~35 31800, V,94,900,9,1, *1~4 31850, V,97,850,21,1, *31~35 32850, V,98,950,21,1, *18~27 3 4500, V,95,850,9,1, *31~35 34850, V,95,950,21,1, *1~4

PAGE 113

113 36700, V,94,800,9,1, *18~27 38250, V,100,950,9,1, *31~35 38350, v,94, 900, 21,1, *1~4 38500, V,95,850,9,1, *31~35 39700, V,97,900,21,1, *18~27 16400, V, 93, 500,9,1, *18~27 16400, V, 98, 450,21,1, 1~4 16400, V, 97, 350,9,1, *31~35 16400, V, 86, 450,9,1, *18~27 16400, V, 95, 300,21,1, *31~35 16400, V, 92, 450,9,1, *18~27 16400, V, 83, 550,21,1, *1~4 1, cross traffic 300, A,45,1000, 6,*1~4 600, A,50,1000, 18,*31~35 900, A,66,1000, 30,*1~4 1 100, A,55,1000, 6,*18~27 1300, A,45,1000, 6,*31~35 1600, A,50,1000, 18,*18~27 1900, A,66,1000, 30,*1~4 2300, A,55,1000, 6,*18~27 2600, A,45,1000, 30,*1~4 3100, A,50,1000, 18,*31~35 3700, A,66,1000, 30,*1~4 4100, A,55,1000, 6,*31~35 4100, A,45,1000 30,*18~27 4600, A,50,1000, 18,*1~4 4650, A,45,900, 6,*18~27 4650, A, 50,900, 18,*18~27 4600, A,45,850, 6,*18~27 4600, A, 50,900, 18,*18~27 4650, A, 53,1000, 30,*18~27 4650, A, 53,1150, 18,*18~27 4700, A, 53,1250, 6,*18~27 4700, A, 53,1400, 6,*18~27 470 0, A, 53,1500, 18,*18~27 4700, A, 53,1300, 30,*18~27 4900, A,66,1000, 30,*31~35 4950, A,55,1000, 6,*1~4 5200, A,45,1000, 30,*18~27 5700, A,50,1000, 18,*1~4 6000, A,66,1000, 30,*31~35 6400, A,55,1000, 6,*31~35 7100, A,45,1000, 30,*1~4 7300, A,50,1000 18,*31~35

PAGE 114

114 7850, A, 53,1000, 30,*1~4 8000, A,66,1000, 30,*18~27 8350, A,55,1000, 6,*31~35 8600, A,45,1000, 30,*1~4 9100, A,50,1000, 18,*18~27 9600, A,66,1000, 30,*18~27 10100, A,55,1000, 6,*31~35 10600, A,45,1000, 30,*1~4 10700, A,50,1000, 18,*18 ~27 11200, A, 53,1000, 30,*1~4 11900, A,66,1000, 30,*31~35 12350, A,55,1000, 6,*31~35 12600, A,45,1000, 30,*1~4 12850, A,50,1000, 18,*31~35 12900, A,66,1000, 6,*18~27 13100, A,55,1000, 6,*1~4 13100, A,45,1000, 18,*31~35 13600, A,50,1000, 18,*31~35 13900, A, 53,1000, 6,*1~4 14400, A,66,1000, 18,*31~35 14750, A,55,1000, 6,*18~27 15300, A,60,1000, 6,*18~27 15300, A,55,850, 18,*18~27 15300, A,58,1200, 6,*18~27 15300, A,55,1400, 18,*18~27 15300, A,53,700, 6,*18~27 1, cross traffic freeway 15850 A,90,1000, 9, *1~35 16000, A,95,1000, 21, *1~35 16200, A,88,1000, 9,*1~4 16850, A,90,1000, 9, *1~35 16900, A,95,1000, 21, *1~35 16950, A,88,1000, 9,*1~35 17450, A,100,1000, 9, *1~35 17700, A,95,1000, 21, *1~4 17900, A,88,1000, 21,*1~35 18250, A, 90,1000, 9, *1~35 18400, A,103,1000, 21, *1~4 19550, A,90,1000, 9, *1~35 20000, A,95,1000, 21, *1~35 20800, A,103,1000, 9,*1~35 21350, A,90,1000, 9, *1~35 21900, A,95,1000, 21, *1~4 22650, A,88,1000, 9,*1~35 23450, A,100,1000, 9, *1~35

PAGE 115

115 24300, A,95 ,1000, 21, *1~4 25100, A,88,1000, 21,*1~35 26250, A,90,1000, 9, *1~35 27000, A,102,1000, 21, *1~4 27950, A,88,1000, 9,*1~35 29000, A,95,1000, 21, *1~4 30800, A,88,1000, 9,*1~35 31350, A,98,1000, 9, *1~35 32900, A,95,1000, 21, *1~35 33650, A,88,100 0, 9,*1~35 34450, A,99,1000, 9, *1~35 35300, A,95,1000, 21, *1~4 36100, A,88,1000, 9,*1~35 37250, A,95,1000, 9, *1~4 38000, A,95,1000, 21, *1~35 38950, A,88,1000, 9,*1~35 1,buildings 1 buildings right side of road 0, BLDG, 50, 80, S1 0, BLDG, 17 0, 80, S3 0, BLDG, 270, 76, S4 0, BLDG, 400, 80, S5 0, BLDG, 450, 80, S6 0, BLDG, 530, 85, S7 0, BLDG, 630, 80, S17 0, BLDG, 730, 80, S9 0, BLDG, 820, 85, G1 0, BLDG, 890, 80, G7 0, BLDG, 930, 75, G8 0, BLDG, 1030, 80, S1 0, BLDG, 1130, 80, S3 0, BLDG, 12 30, 76, S4 0, BLDG, 1330, 80, S5 0, BLDG, 1430, 80, S6 1 first intersection 0, BLDG, 1700, 80, S1 0, BLDG, 1820, 80, S3 0, BLDG, 1920, 76, S4 0, BLDG, 2050, 80, S5 0, BLDG, 2100, 80, S6 0, BLDG, 2180, 85, S7 0, BLDG, 2280, 80, S17 0, BLDG, 2380, 80, S9 0, BLDG, 2470, 85, G1

PAGE 116

116 0, BLDG, 2540, 80, G7 0, BLDG, 2580, 75, G8 0, BLDG, 2680, 80, S1 0, BLDG, 2780, 80, S3 0, BLDG, 2880, 76, S4 0, BLDG, 2980, 80, S5 0, BLDG, 3080, 80, S6 0, BLDG, 3180, 80, S1 0, BLDG, 3300, 80, S3 0, BLDG, 3400, 76, S4 0, BLDG, 35 30, 80, S5 0, BLDG, 3580, 80, S6 0, BLDG, 3660, 85, S7 0, BLDG, 3760, 80, S17 0, BLDG, 3860, 80, S9 0, BLDG, 3950, 85, G1 0, BLDG, 4020, 80, G7 0, BLDG, 4060, 75, G8 0, BLDG, 4160, 80, S1 0, BLDG, 4260, 80, S3 0, BLDG, 4360, 76, S4 0, BLDG, 4460, 80, S5 0 BLDG, 4560, 80, S6 0, BLDG, 4660, 80, S7 0, BLDG, 4900, 80, S3 0, BLDG, 5000, 76, S4 0, BLDG, 5130, 80, S5 0, BLDG, 5280, 80, S6 0, BLDG, 5360, 85, S7 0, BLDG, 5460, 80, S17 0, BLDG, 5560, 80, S9 0, BLDG, 5650, 85, G1 0, BLDG, 5720, 80, G7 0, BLDG, 5860, 75, G8 1 second intersection 0, BLDG, 4880, 180, S11, 90 0, BLDG, 4880, 300, S13, 90 0, BLDG, 4880, 400, S14, 90 0, BLDG, 4880, 530, S15, 90 0, BLDG, 4880, 780, S16, 90 0, BLDG, 4880, 860, S17, 90 0, BLDG, 4880, 960, S7, 90 0, BLDG, 4880, 1060, S10, 90 0, BLDG, 4880, 1150, G8, 90 0, BLDG, 4880, 1220, G7, 90

PAGE 117

117 0, BLDG, 4880, 1260, G1, 90 0, BLDG, 4880, 1360, S1, 90 0, BLDG, 4880, 1460, S2, 90 0, BLDG, 4880, 1560, S12, 90 0, BLDG, 4880, 1660, S3, 90 0, BLDG, 4880, 1760, S5, 90 0, BLDG, 4880, 1860, S6, 90 0, BLDG, 4880, 1980, S4, 90 0, BLDG, 4880, 2065, S6, 90 0, BLDG, 4880, 2155, S11, 90 0, BLDG, 4880, 2235, G1, 90 0, BLDG, 4880, 2335, G7, 90 0, BLDG, 4880, 2440, G11, 90 0, BLDG, 4880, 2530, S11, 90 0, BLDG, 4880, 2650, S2, 90 0, BLDG 4880, 2750, S4, 90 0, BLDG, 4880, 2880, S6, 90 0, BLDG, 4880, 3150, S9, 90 0, BLDG, 4880, 3250, G1, 90 0, BLDG, 4880, 3350, G7, 90 0, BLDG, 4880, 3450, G8, 90 0, BLDG, 4880, 3550, S1, 90 0, BLDG, 4880, 3650, S3, 90 0, BLDG, 4880, 3750, S4, 90 0, BLDG, 4880, 3850, S5, 90 0, BLDG, 4880, 3950, S6, 90 1 third intersection 0, BLDG, 5000, 2930, S1, 180 0, BLDG, 5090, 2930, S3, 180 0, BLDG, 5200, 2930, S4, 180 0, BLDG, 5300, 2930, S5, 180 0, BLDG, 5400, 2930, S6, 180 0, BLDG, 5500, 2930, S7, 180 0, BLDG, 5600, 2930, S17, 180 0, BLDG, 5700, 2930, S9, 180 0, BLDG, 5800, 2930, G1, 180 0, BLDG, 5900, 2930, G7, 180 0, BLDG, 6000, 2930, G8, 180 0, BLDG, 6100, 2930, S1, 180 0, BLDG, 6200, 2930, S3, 180 0, BLDG, 6300, 2930, S4, 180 0, BLDG, 64 00, 2930, S5, 180 0, BLDG, 6500, 2930, S6, 180 0, BLDG, 6600, 2930, S4, 180

PAGE 118

118 0, BLDG, 6700, 2930, S5, 180 0, BLDG, 6800, 2930, S1, 180 0, BLDG, 6890, 2930, S3, 180 0, BLDG, 7000, 2930, S4, 180 0, BLDG, 7100, 2930, S5, 180 0, BLDG, 7200, 2930, S6, 1 80 0, BLDG, 7300, 2930, S7, 180 0, BLDG, 7400, 2930, S17, 180 0, BLDG, 7500, 2930, S9, 180 0, BLDG, 7600, 2930, G1, 180 0, BLDG, 7700, 2930, G7, 180 0, BLDG, 7900, 2930, S1, 180 0, BLDG, 8000, 2930, S3, 180 0, BLDG, 8100, 2930, S2, 180 0, BLDG, 820 0, 2930, S1, 180 0, BLDG, 8290, 2930, S3, 180 0, BLDG, 8400, 2930, S4, 180 0, BLDG, 8500, 2930, S5, 180 0, BLDG, 8600, 2930, S6, 180 0, BLDG, 8700, 2930, S7, 180 0, BLDG, 8800, 2930, S17, 180 0, BLDG, 8900, 2930, S9, 180 0, BLDG, 9000, 2930, G1, 1 80 0, BLDG, 9100, 2930, G7, 180 0, BLDG, 9200, 2930, G8, 180 0, BLDG, 9300, 2930, S1, 180 0, BLDG, 9400, 2930, S3, 180 0, BLDG, 9500, 2930, S4, 180 1 forth intersection 0, BLDG, 9700, 2800, H1, 270 0, BLDG, 9700, 2600, B1, 270 0, BLDG, 9700, 2400 H2, 270 0, BLDG, 9700, 2000, H3, 270 0, BLDG, 9700, 1800, H4, 270 0, BLDG, 9700, 1600, H6, 270 0, BLDG, 9700, 1300, H5, 270 0, BLDG, 9700, 1000, H3, 270 1 buildings of left side of the road 0, BLDG, 50, 80, S10 0, BLDG, 150, 80, S11 0, BLDG, 250 83, S12 0, BLDG, 350, 80, S13 0, BLDG, 450, 88, S15

PAGE 119

119 0, BLDG, 550, 80, S16 0, BLDG, 670, 80, S14 0, BLDG, 755, 80, S8 0, BLDG, 844, 80, S1 0, BLDG, 924, 80, G3 0, BLDG, 1030, 80, S1 0, BLDG, 1130, 80, S3 0, BLDG, 1230, 76, S4 0, BLDG, 1330, 80, S5 0, BLDG, 1430, 80, S6 1 first intersection 0, BLDG, 1700, 80, S10 0, BLDG, 1750, 80, S11 0, BLDG, 1850, 83, S12 0, BLDG, 1950, 80, S13 0, BLDG, 2050, 88, S15 0, BLDG, 2150, 80, S16 0, BLDG, 2270, 80, S14 0, BLDG, 2355, 80, S8 0, BLDG, 2444, 80, S1 0, BLDG 2524, 80, G3 0, BLDG, 2630, 80, S1 0, BLDG, 2730, 80, S3 0, BLDG, 2830, 76, S4 0, BLDG, 2930, 80, S5 0, BLDG, 3030, 80, S6 0, BLDG, 3130, 80, S10 0, BLDG, 3230, 80, S11 0, BLDG, 3330, 83, S12 0, BLDG, 3430, 80, S13 0, BLDG, 3530, 88, S15 0, BLDG, 3 630, 80, S16 0, BLDG, 3750, 80, S14 0, BLDG, 3835, 80, S8 0, BLDG, 3924, 80, S1 0, BLDG, 4004, 80, G3 0, BLDG, 4110, 80, S1 0, BLDG, 4210, 80, S3 0, BLDG, 4310, 76, S4 0, BLDG, 4410, 80, S5 0, BLDG, 4510, 80, S6 0, BLDG, 4600, 80, S10 0, BLDG, 4650, 80, S11 0, BLDG, 4930, 80, S13 0, BLDG, 5030, 88, S15

PAGE 120

120 0, BLDG, 5130, 80, S16 0, BLDG, 5250, 80, S14 0, BLDG, 5335, 80, S8 0, BLDG, 5424, 80, S1 0, BLDG, 5504, 80, G3 0, BLDG, 5610, 80, S1 0, BLDG, 5710, 80, S3 0, BLDG, 5810, 76, S4 1 second intersecti on 0, BLDG, 4720, 580, S6, 90 0, BLDG, 4720, 180, S1, 90 0, BLDG, 4720, 300, S3, 90 0, BLDG, 4720, 400, S4, 90 0, BLDG, 4720, 530, S5, 90 0, BLDG, 4720, 780, S6, 90 0, BLDG, 4720, 860, S7, 90 0, BLDG, 4720, 960, S17, 90 0, BLDG, 4720, 1060 S9, 90 0, BLDG, 4720, 1150, G1, 90 0, BLDG, 4720, 1220, G7, 90 0, BLDG, 4720, 1260, G8, 90 0, BLDG, 4720, 1360, S10, 90 0, BLDG, 4720, 1460, S11, 90 0, BLDG, 4720, 1560, S12, 90 0, BLDG, 4720, 1660, S13, 90 0, BLDG, 4720, 1760, S15, 90 0 BLDG, 4720, 1860, S16, 90 0, BLDG, 4720, 1980, S14, 90 0, BLDG, 4720, 2065, S8, 90 0, BLDG, 4720, 2155, S1, 90 0, BLDG, 4720, 2235, G3, 90 0, BLDG, 4720, 2335, G8, 90 0, BLDG, 4720, 2440, G11, 90 0, BLDG, 4720, 2530, S1, 90 0, BLDG, 4720, 2650, S3, 90 0, BLDG, 4720, 2750, S4, 90 0, BLDG, 4720, 2880, S5, 90 0, BLDG, 4720, 3150, S1, 90 0, BLDG, 4720, 3250, G3, 90 0, BLDG, 4720, 3350, S1, 90 0, BLDG, 4720, 3450, S3, 90 0, BLDG, 4720, 3550, S4, 90 0, BLDG, 4720, 3650, S5, 90 0, BLDG, 4720, 3750, S6, 90 0, BLDG, 4720, 3850, S10, 90

PAGE 121

121 0, BLDG, 4720, 3950, S11, 90 1 third intersection 0, BLDG, 5000, 3100, S6 0, BLDG, 5090, 3100, S17 0, BLDG, 5200, 3100, S9 0, BLDG, 5300, 3100, G1 0, BLDG, 5400, 3100, G7 0, BLDG, 5500, 3100, G8 0, BLDG, 5600, 3100, S1 0, BLDG, 5700, 3100, S3 0, BLDG, 5800, 3100, S4 0, BLDG, 5900, 3100, S5 0, BLDG, 6000, 3100, S6 0, BLDG, 6100, 3100, S4 0, BLDG, 6200, 3100, S5 0, BLDG, 6300, 3100, S1 0, BLDG, 6400, 3100, S3 0, BLDG, 6500, 3100, S4 0, BLDG, 6600, 3100, S5 0, BLDG, 6700, 3100, S6 0, BLDG, 6800, 3100, S7 0, BLDG, 6890, 3100, S17 0, BLDG, 7000, 3100, S9 0, BLDG, 7100, 3100, G1 0, BLDG, 7200, 3100, G7 0, BLDG, 7300, 3100, G8 0, BLDG, 7400, 3100, S1 0, BLDG, 7500, 3100, S3 0 BLDG, 7600, 3100, S4 0, BLDG, 7700, 3100, S5 0, BLDG, 7900, 3100, S3 0, BLDG, 8000, 3100, S4 0, BLDG, 8100, 3100, S5 0, BLDG, 8200, 3100, S6 0, BLDG, 8290, 3100, S17 0, BLDG, 8300, 3100, S9 0, BLDG, 8400, 3100, G1 0, BLDG, 8500, 3100, G7 0, BLD G, 8600, 3100, G8 0, BLDG, 8700, 3100, S1 0, BLDG, 8800, 3100, S3 0, BLDG, 8900, 3100, S4 0, BLDG, 9000, 3100, S5 0, BLDG, 9100, 3100, S6 0, BLDG, 9200, 3100, S4

PAGE 122

122 0, BLDG, 9300, 3100, S5 0, BLDG, 9400, 3100, S1 0, BLDG, 9500, 3100, S3 1 forth in tersection 0, BLDG, 9900, 2800, H6, 90 0, BLDG, 9900, 2600, H3, 90 0, BLDG, 9900, 2300, H2, 90 0, BLDG, 9900, 2000, B1, 90 0, BLDG, 9900, 1700, H5, 90 0, BLDG, 9900, 1500, H4, 90 0, BLDG, 9900, 1200, H2, 90 0, BLDG, 9900, 900, B3, 90 1 trees 0, TREE, 50, 0, 1, 60, 61, 0 35300, ES

PAGE 123

123 APPENDIX C QUESTIONNAIRES Informed Consent Form Protocol Title : Driver Assistance Systems in a Driving Simulator Please read this consent document c arefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to capture important changes in driving behavior caused by advanced driver assistance systems. These changes could affect traffic and even help in mitigating congestion. What you will be asked to do in the study: in a driving simulator for about 15 minutes each scenario. Upon arrival and before starting t he experiment, a check in procedure will be undertaken as follows: 1) sign the informed consent form (this form), 2) complete the background survey form, 3) provide Questionnaire) and be subjected to an acclimation in the driving simulator, 5) answer the SSQ again if no sign of symptoms is observed, continue with the experiment. Then you will be subjected to the first scenario, simulating a trip in a freeway, and an arterial, in a n ormal vehicle, without any additional features. After this, you will be explained how the two tested assistance systems work, and you will be subjected to the second scenario, that is similar to the first one, but simulating a vehicle implemented with the two systems. During data collection, you will not be interrupted. After the completion of each scenario, you will be asked to answer a questionnaire regarding your experience using the two assistance systems, and the SSQ again. The data collected in the ex periment, including both the driving simulator scenarios and the information obtained from the questionnaires, will be used for traffic engineering research only. Your identity will not be revealed in the final manuscript. The time planned for this activit y is 75 minutes, including time for questionnaires before and after the completion of each scenario. Time Required: Up to 1 hour and 15 minutes Risks and Benefits:

PAGE 124

124 The risk for this experiment will be related to driving simulator sickness. Some users can experiment adverse effects, including nauseam, oculomotor discomfort, and disorientation. You will be tested for simulator sickness before the experiment, and if you experience any of this, the simulation will be discontinued. You will be accompanied by th e principal investigator (Barbara Martin), and be instructed to drive the simulator as you usually do during the entire data collection. We do not anticipate that you will benefit directly by participating in this experiment. Compensation: You will receiv e a $50 compensation for participating in this field d ata collection experiment. If you withdraw after completing one of the two scenarios, you will receive $25 for compensation. No compensation will be paid for withdrawing before the completion of the fir st scenario. Confidentiality: Information collected from this experiment will be used for traffic engineering research only. Your identity will be kept confidential to the extent provided by law. In accordance with the Confidential Information Protection and Statistical Efficiency Act of 2002 (Title 5 of Public Law 107 347) and other applicable Federal laws, your responses will not be disclosed in identifiable form without your consent. Voluntary participation: Your participation in this study is complete ly voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from the study at anytime without consequence. Whom to contact if you have questions about the study: Barbara Martin, Graduate Stu dent, Department of Civil and Coast Engineering, Room 518, Weil hall, Phone: (352)392 9537 x1541. Lily Elefteriadou, Ph.D., Department of Civil and Coast Engineering, Room 512, Weil hall, Phone: (352)392 9537 x1452. Whom to contact about your rights as a research participant in the study: UFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611 2250, Phone: (352)392 0433. Agreement:

PAGE 125

125 I have read the procedure described above. I voluntarily agree to participate in the procedure and I have r eceived a copy of this description. Pa rticipant:_________________________ ____________ Date: _________________ Principal Investigator:____________________ _________ Date: _________________

PAGE 126

126 Transportation Research Center Prescreening Questionnai re for Driving Behavior Change with Advanced Drivers Assistance Systems in a Driving Simulator Research To Participants: This questionnaire is used to select a diverse pool of drivers to participate in the data collection experiment Information collected in this form will be used for traffic engineering research only. All responses will be held in complete confidential and exempt ed from public disclosure by law In accordance with the Confidential Information Protection and Statistical Efficien cy Act of 2 002 (Title 5 of Public Law 107 347) and other applicable Federal laws, your responses will not be disclosed in identifiable form without your consent. Since driver diversities are highly encouraged, only the most fitful responder s will be chosen Please answer as many as possible. Return Address: By Email : barbaramartin @ufl.edu By M ail : Barbara Martin 511 Weil Hall, PO Box 116580 Gainesville, FL 32611 1) What is your gender? Male Female 2) What is your a ge range? less than 25 years 25 to 30 years 31 to 40 years 41 to 50 years 51 to 60 years 61 years or more 3) Which of the following groups do you most identify yourself as ? Caucasian Native American African American Hispanic Asian Pacific Islander Other (please specify) 4) Wh ere did you begin your driving practice and obtained your driver license ? North America Latin America Asia Europe Australia

PAGE 127

127 Other (please specify) 5) How long have you been driving in the U.S.? less than 1 year 1 to 3 years 3 to 9 years more than 10 years 6) Yes No 7) What is your occu pation? Full time student University faculty/staff Professional driver Other (please specify) 8) How often d o you drive to work/school? everyday usually sometimes never 9) How much time do you spend driving per week? less than 4 hr 4 to 8 hr 8 to 14 hr more than 14 hr 10) What time of the day do you usually drive ? AM/PM peak hour (6am 10am; 4pm 7pm) during work days Non peak hours (including holiday and weekend) 11) What ty pe of vehicle do you usually drive ? Sedan/Coupe Pickup/SUV Jeep Truck 12) Do you have any health problems? Yes (please specify) No 13) Do you use a pace maker? Yes No 14) Will you be willing to respond a questionnaire to assess your cognitive function?

PAGE 128

128 Yes No 15) What time are you typically ava ilable for participating in the experiments ? Please check as many as possible. Monday morning (9:00am to 12:00pm) Tuesday morning (9:00am to 12:00pm) Wednesday morning (9:00am to 12:00pm) Thursday morning (9:00am to 12:00pm) Friday morning (9:00am to 12:00pm) Monday afternoon (1:00pm to 4:00pm) Tuesday afternoon (1:00pm to 4:00pm) Wednesday afternoon (1:00pm to 4:00 pm) Thursday afternoon (1:00pm to 4:00pm) Friday afternoon (1:00pm to 4:00pm) Weekends only Any time by appointment (at least 1 from phone/email/mail ) Name (Required) Email Mail Address Phone Date

PAGE 129

129 Transportation Research Center Mini Mental State Examination (MMSE) Date : _____________ Orientation 1. (i) What day of the week is it? Day Date Month Year (iii) What season are we in? Season 2. Wh ere are you? City County State Hospital/House Floor Registration 3 say the names of 3 unrelated objects Apple Please lis ten carefully and try to remember them. Table When I am done, repeat the 3 names. Penny Attention/Calculation 4. D L R O W Recall 5 Please repeat the 3 words I said before. Apple Table Penny Language 6 Show the person a wristwatch and Watch ask him/ her what it is. Repeat for pencil. Pencil 7. No mistakes 8. Follow a 3 stage command: fold it in h alf, and

PAGE 130

130 9. Read and obey the following: Close your eyes. Write a sentence. Copy the following design.

PAGE 131

131 Transportation Re search Center Simulator Sickness Questionnaire : ____________________ Date : ______________ Tick the box that best represents how you are feeling regarding each symptom: General discomfort none s light moderate severe Fatigue none slight moderate severe Headache none slight moderate severe Eyestrain none slight moderate severe Difficulty focusing none slight moderate severe Increased salivation none slight moderate severe Sweating none slight moderate severe Nausea none slight moderate severe Difficulty concentrating none slight moderate severe Fullness of head none slight moderate severe Blurred vision none slight moderate severe Dizzy (eyes open) none slight moderate severe

PAGE 132

132 Dizzy (eyes closed) none slight moderate severe Vertigo none slight moderate severe Stomach awareness none slight moderate severe Burping none slight moderate severe

PAGE 133

133 Transportation Research Center : ____________________________ Date : ______________ Note: Information collected in this form will be used for traffic engineerin g research only. All responses will be held in complete confidential and exempt from public disclosure by law In accordance with the Confidential Information Protection and Statistical Efficien cy Act of 2002 (Title 5 of Public Law 107 347) and other appli cable Federal laws, your responses will not be disclosed in identifiable form without your consent. By law, every interviewer, as well as every agent, is subject to a jail term, a fine, or both if he or she makes public ANY identifiable information you rep orted. 16) If the speed limit on the freeway is 7 0 mph, what speed are you likely to drive (assuming good visibility and good weather conditions) ? less than 65 mph 65 to 70 mph 70 to 75 mph more than 80 mph 17) How often do you change lanes if the vehicle in front of you is slower? Very often Sometimes Seldom 18) What type of driver do you consider yourself ? Very aggressiv e Somewhat aggressive Somewhat conservative Very conservative 19) What type of driver do your friends and family consider you ? Very aggressive Somewhat aggressive Somewhat conservative Very conservative 20) When planning your driving trip, d o you allow additional time for possible delays due to congestion construction or bad weather? Yes, always Sometimes Never

PAGE 134

134 Transportation Research Center User Acceptance Questionnaire : _________________________ Date : _____________ My judgments of the Adaptive Cruise Control System ox on every line) 1 useful useless 2 pleasant unpleasant 3 bad good 4 nice annoying 5 effective superfluous 6 irritating likeable 7 assisting worthless 8 undesirable desirable 9 raising alertness sleep inducing

PAGE 135

135 My judgments of the Lane Change Assist System every line) 1 useful us eless 2 pleasant unpleasant 3 bad good 4 nice annoying 5 effective superfluous 6 irritating likeable 7 assisting worthless 8 undesirable desirable 9 raising alertness sleep inducing My judgments of the combination of the two Assistance Systems tick a box on every line) 1 useful useless 2 pleasa nt unpleasant 3 bad good 4 nice annoying 5 effective superfluous 6 irritating likeable 7 assisting worthless 8 undesirable desirable 9 raising alertness sleep inducing

PAGE 136

136 LIST OF REFERENCES AAA Foundation for Traffic Safety. (2009). Aggressive driving: research update. Washington, D.C. A new gene Auto Technology 5(Feb.), 62 63. Allen, R. W., Park, G. D., Fiorentino, D., Rosenthal, T. J., and Cook, L. M. (2006). 9th Annual Driving Simulation Confere nce Europe Paris, France. Auckland, R. A., Manning, W. J., Carsten, O. M. J., and Jamson, A. H. (2008). Vehicle System Dynamics 46(Suppl. 1), 883. Blana, E. (1996). A survey of driving research simulators around the world ITS Working Paper 481, Institute for Transport Studies, University of Leeds, Leeds, UK. IE EE Transactions on Intelligent Transportation Systems 4(4), 173 188. Advanced Driver Assistance Systems European Journal of Transport and Infrastructure Rese arch 1(3), 245 253. Vehicle System Dynamics 44(3), 223 245. Center ed Design of an Acc With Braking and Forward Crash Vehicle System Dynamics 36(2), 203. UMTRI Research Review 29(4), 1 17. Fancher, P., Ervin, R., Saye r, J., Hagan, M., Bogard, S., Bareket, Z., Mefford, M., and Haugen, J. (1998). Intelligent cruise control field operational test Interim Report, University of Michigan Transportation Research Institute, Ann Arbor, MI. ed Vehicle Proceedings of the ITS World Congress London, UK.

PAGE 137

137 mental state" : A practical method for grading the cognitive state of patients for the cli Journal of Psychiatric Research 12(3), 189 198. Accident analysis & prevention 34(5), 589 600. Golias, J., Yannis, G., and Antoniou, C. assistance Transport Reviews 22(2), 179 196. cost, multiple driver in the IEEE Control Systems Magazine 26(3), 42 55. IEEE Intelligent Transportation Systems Conference Proceedings 506 509. Ioannou, P., and Chi IEEE Transactions on Vehicular Technology 42(4), 657 672. IEEE Tr ansactions on Intelligent Transportation Systems 6(1), 79 89. Sickness Questionnaire: An Enhanced Method for Quantifying Simulator The International Journal of Aviation Psychology 3(3), 203. Temporal Congested Traffic Patterns at Arxiv preprint cond mat/0309017.[Online]. Available:< http://arxiv. org/abs/cond mat/0309017 > reakdown probability model at freeway ramp merges based on < http://www.umtri.umich.edu/expertiseSub.php?esID=61 > (Sep. 22, 2010). L eBlanc, D., Bezzina, D., Tiernan, T., Freeman, K., Gabel, M., and Pomerleau, D. (2008). System performance guidelines for a prototype integrated vehicle based safety system (IVBSS) light vehicle platform. The University of Michigan Transportation Research Institute, Ann Arbor, MI.

PAGE 138

138 Transportation Research Record: Journal of the Transportation Research Board 1980(1), 49 5 6. IEEE International Conference on Robotics and Automation vehic JSME International Journal Series C 43(3), 671 677. gap settings of adaptive cruise control (ACC) on driving performance and subjective acceptance in a bus Safety Sci ence 47(5), 620 625. Progress in Technology SAE International 70, 205 248. Mezny, B., Laborczi, P., and hoc adaptive cruise control Infocommunications Journal Selected Papers Vol. 64(1). Human Factors and Ergo nomics Society Annual Meeting Proceedings 534 537. http://driving.phhp.ufl.edu/virtual > (Sep. 22, 2010). Internati onal Joint Conference, SICE ICASE. 2138 2141. Transport Reviews: A Transnational Transdisciplinary Journal 28(5), 659. Reynaldo, J., and Sa Extension Information Technology 37(2). Proceedings of the 12 th IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications IEEE Computer Society, Washington, D.C., 217 226. IEEE Intelligent Vehicle Symposi um 240 244.

PAGE 139

139 IEEE International Conference on Systems, Man and Cybernetics 288 293. International Journal of Human Computer Studies 65(3), 192 205. Critical Driving Errors of On The American Journal of Occupational Therapy 64(2), 242 251. errors between on the road and simulated driving ass essment: A validation Traffic Injury Prevention 10(4), 379 385. by wire: The case of driver Safety Science 27(2 3), 149 15 9. Accident Analysis & Prevention 31(5), 567 578. ay Traffic Including Automatisierungstechnik 49(11), 478 484. Transportation Research Part C: Emerging Technologies 5(1), 1 10. Transportation Research Record: Journal of the Transportation Re search Board 1800(1), 78 84. Proceedings of the 15th Triennial IFAC World Conference Barcelona, Spain. Y a mixed traffic system consisting of ACC vehicles and manual vehicles: A hybrid Physica A: Statistical Mechanics and its Applications 388(12), 2 483 2491.

PAGE 140

140 BIOGRAPHICAL SKETCH Brbara Barq ueta Martin earned her B.S. in c ivil e ngineering from the University of transportation engineering at the University of Flor ida and graduated in December 2010. She intends to work on transportation projects related to the 2014 FIFA World Cup and the 2016 Summer Olympics, both hosted by cities in Brazil.