Modelling Driver Behaviour in Automotive Environments - Critical Issues in Driver Interactions with Intelligent Transport Systems

Title Page

Copyright Page

Table of Contents

Editorial

List of Contributors

I International Projects and Actions on Driver Modelling

1 Modelling Driver Behaviour in European Union and International Projects

1.1 Introduction

1.2 Evaluation of Driver Behaviour Models

1.2.1 Michon's Hierarchical Control Model

1.2.2 The GADGET-Matrix: Integrating Hierarchical Control Models and Motivational Models of Driver Behaviour

1.2.3 DRIVABILITY Model

1.3 Driver Behaviour Adaptation Models and Their Relation to ADAS

1.3.1 Automaticity

1.3.2 Locus of Control

1.3.3 Risk Homeostasis

1.3.4 Risk Compensation

1.3.5 Threat Avoidance

1.3.6 Utility Maximisation

1.3.7 Behavioural Adaptation Formula

1.4 Use of Driver Behaviour Models in EU and International Projects

1.4.1 Driver Models Use for Driver Training and Assessment

1.4.2 Evaluation ofDriver Models' Use for Safety Aids

1.4.2.1 Use of Seat Belts

1.4.2.2 Use of Motorcycle Helmet

1.4.2.3 Studded Tyres

1.4.2.4 Antilock Braking Systems

1.4.3 Driver Models Use for ADAS Design and Impact Assessment

1.5 Conclusions

References

2 TRB Workshop on Driver Models: A Step Towards a Comprehensive Model of Driving?

2.1 Introduction

2.2 Workshop Presentation and Speakers' Contribution

2.2.1 Workshop Content

2.2.1.1 Driver Model Purpose and Application

2.2.1.2 Driver Model Architecture and Implementations

2.2.1.3 Calibration and Validation

2.2.2 Summaries ofthe Speakers' Contributions

2.2.2.1 In-Vehicle Information System - Jon Hankey

2.2.2.2 ACT-R Driver Model- Dario Salvucci

2.2.2.3 Optimal Control Model - Richard van der Horst

2.2.2.4 ACME

2.2.2.5 Fuzzy Logic Based Motorway Simulation

2.3 Synthesis of Presented Models

2.3.1 Understanding Models' Scope

2.3.2 Driver Model Toolbox

2.4 Towards a Comprehensive Model of Driving

2.5 Conclusions

References

3 Towards Monitoring and Modelling for Situation-Adaptive Driver Assist Systems

3.1 Introduction

3.2 Behaviour-Based Human Environment Creation Technology Project

3.2.1 Aims of the Project

3.2.2 Measurement of Driving Behaviour

3.2.3 Driving Behaviour Modelling

3.2.4 Detection of Non-Normative Behaviour

3.2.5 Estimation of Driver's State

3.2.5.1 Estimation of Driver 's Mental Tension

3.2.5.2 Estimation of Driver's Fatigue

3.3 Situation and Intention Recognition for Risk Finding and Avoidance Project

3.3.1 Aims of the Project

3.3.2 Adaptive Function Allocation Between Drivers and Automation

3.3.3 Decision Authority and the Levels of Automation

3.3.4 Model-Based Evaluation of Levels of Automation

3.3.4.1 Drivers' Psychological States and Their Transitions

3.3.4.2 Driver's Response to an Alert

3.3.4.3 Evaluation of Efficacy of Levels of Automation

3.4 Concluding Remarks

References

II Conceptual Framework and Modelling Architectures

4 A General Conceptual Framework for Modelling Behavioural Effects of Driver Support Functions

4.1 Introduction

4.2 Intended Application Areas and Requirements

4.2.1 Functional Characterisation of Driver Support Functions

4.2.2 Coherent Description ofExpected Behavioural Effects of Driver Support Functions

4.2.3 Conceptualising Relations Between Behavioural Effects and Road Safety

4.2.4 Specific Requirements

4.3 Existing Models of Driver Behaviour

4.3.1 Manual Control Models

4.3.2 Information Processing Models

4.3.3 Motivational Models

4.3.4 Safety Margins

4.3.5 Hierarchical Models

4.4 A Conceptual Framework

4.4.1 Driver Behaviour as Goal-Directed Activity

4.4.2 Dynamical Representation of Driver Behaviour

4.4.3 The Contextual Control Model (COCOM)

4.4.4 The Extended Control Model (ECOM)

4.5 Application

4.5.1 Characterising Driver Support Functions

4.5.1.1 Support for Tracking

4.5.1.2 Support for Regulating

4.5.1.3 Support for Monitoring

4.5.1.4 Support for Targeting

4.5.1.5 Non-Driving-Related Functions

4.5.1.6 Workload Management Functions

4.5.2 Characterising Behavioural Effects of Driver Support Functions

4.5.2.1 Behavioural Adaptation to Driving Support Functions

4.5.2.2 Effects of Multitasking While Driving

4.5.3 Driver Behaviour and Accident Risk

4.6 Discussion and Conclusions

References

5 Modelling the Driver in Control

5.1 Introduction

5.2 A Cognitive View of Driving

5.3 Human Abilities

5.4 Classifying Driver Behaviour Models

5.5 Hierarchical Control Models

5.6 Control Theory

5.7 Adaptive Control Models

5.8 Cognition in Control

5.9 Goals for Control

5.10 Time and Time Again

5.11 Multiple Layers of Control

5.12 Joystick Controlled Cars - An Example

5.13 Summary and Conclusion

References

6 From Driver Models to Modelling the Driver: What Do We Really Need to Know About the Driver?

6.1 Introduction

6.2 A Typology of Models

6.3 Descriptive Models

6.3.1 Task Models

6.3.2 Adaptive Control Models

6.3.3 Production Models

6.4 Motivational Models

6.5 Towards a Real-Time Model of the Driver

6.5.1 What Type of Model Is Required?

6.5.2 Grouping the Factors

6.5.3 A Proposed Structure

6.5.4 Verifying the Model

6.6 Developing an Online Model

6.7 Conclusions

References

III Learning and Behavioural Adaptation

7 Subject Testing for Evaluation of Driver Information Systems and Driver Assistance Systems - Learning Effects and Methodological Solutions

SUMMARY

7.1 Introduction

7.2 Methodological Issues

7.3 Experimental Examples

7.3.1 Evaluation of a Multimodal HMI

7.3.2 Destination Entry While Driving

7.3.3 Evaluation of Driver Assistance Systems

7.4 Solutions

7.5 Conclusions

References

8 Modelling Driver's Risk Taking Behaviour

8.1 Introduction

8.2 Expected Risk Reductions from New Technology on the Road

8.3 Behaviour When Driving with Supports

8.3.1 The Importance of Plain Old Ergonomics

8.3.2 The Loss of Potentially Useful Skills

8.3.3 Opportunities for New Errors

8.3.4 Problematic Transitions

8.3.5 Risk and Risk Perception: My Risk and Yours

8.4 Behavioural Adaptation

8.4.1 Direct Changes in Behaviour

8.4.2 A Word of Caution About Working with Risk Measures in Traffic Safety Studies

8.4.3 A Piece of Empirical Evidence from Seat Belt Accident Statistics

8.4.4 Higher-Order Forms ofAdaptation

8.5 The Link Between Behaviour and Risk

8.5.1 Average Speed, Speed Variability and Risk

8.5.2 Lane-Keeping Performance and Risk

8.5.3 Car-Following and Risk

8.6 Countermeasures Against Behavioural Adaptation

8.6.1 Should There Be Any?

8.6.2 Incentive Schemes and Their Expected Results

8.7 Conclusions

8.8 An Afterthought

References

9 Dealing with Behavioural Adaptations to Advanced Driver Support Systems

9.1 Introduction

9.2 'Behavioural Adaptation' in Road Safety Research

9.3 Behavioural Adaptation to Advanced Driver Support Systems

9.3.1 The Diversity of Behavioural Changes Studied and Observed

9.3.2 The Importance of the Situational Context and the Interactive Dimension of Driving

9.3.3 The Potential Differential Impact of Driver Support Systems

9.3.4 Learning to Drive with New Driver Support Systems

9.4 Behavioural Adaptation in the AIDE Project

References

IV Modelling Motivation and Psychological Mechanisms

10 Motivational Determinants of Control in the Driving Task

10.1 Introduction

10.2 Understanding Speed Choice

10.2.1 Behaviour Analysis

10.2.2 The Theory of Planned Behaviour

10.2.3 Risk Homeostasis Theory

10.2.4 The Task-Capability Interface Model

10.2.4.1 The Determination of Task-Difficulty Level: Task-Difficulty Homeostasis

10.2.4.2 The Representation of Task-Difficulty

10.2.5 The Somatic-Marker Hypothesis

10.2.5.1 Predictions and Speculations from the Somatic-Marker Hypothesis

10.3 Conclusions

References

11 Towards Understanding Motivational and Emotional Factors in Driver Behaviour: Comfort Through Satisficing

11.1 Introduction

11.2 Emotional Tension and 'Risk Monitor'

11.3 Safety Margins and Safety Zone

11.4 Available Time, Workload and Multilevel Task Control

11.5 Safety Margins, Affordances and Skills

11.6 Towards Unifying Emotional Conceptsin Routine Driving

11.6.1 Safety Margins - To Control and Survive

11.6.2 Vehicle/Road System - To Provide Smooth and Comfortable Travel

11.6.3 Rule Following - ToAvoid Sanctions

11.6.4 Good (or Expected) Progress of Trip -Mobility and Pace/Progress

11.7 Comfort Through Satisficing

11.8 'Go to the Road': Need of On-Road Research

References

12 Modelling Driver Behaviour on Basis of Emotions and Feelings: Intelligent Transport Systems and Behavioural Adaptations

12.1 Introduction

12.2 Defining Motivation

12.3 Motivational Aspects in Driver Behaviour Models

12.4 Behavioural Adaptation and Risk Compensation

12.5 Wilde's Risk Homeostasis Theory (RHT)

12.5.1 Target Risk or Target Feeling?

12.6 Effects of ABS: An Illustrative Example of ITS

12.7 Issues Raised by the Example of ABS: The Relevance for ITS

12.8 Adaptation - Mismatch Between Technology and Human Capability

12.9 ITS Technology May Enhance As Well As Reduce the Window of Opportunities

12.10 Damasio and the Somatic Marker Hypothesis

12.11 The Monitor Model

12.12 The Monitor Model and Prediction of Effects of ITS

12.13 Summary and Conclusions

References

V Modelling Risk and Errors

13 Time-Related Measures for Modelling Risk in Driver Behaviour

13.1 Introduction

13.2 The Driving Task

13.3 Lateral Control

13.3.1 Time-to-Line Crossing (TLC)

13.3.2 Lateral Distance When Passing

13.4 Longitudinal Control

13.4.1 Time-to-Collision (ITC)

13.4.2 Time-to-Intersection (TTl)

13.4.3 Time-to-Stop-Line (ITS)

13.5 Conclusions

References

14 Situation Awareness and Driving: A Cognitive Model

14.1 Introduction

14.2 Situation Awareness

14.2.1 An Algorithmic Description of Situation Awareness

14.2.1.1 The Construction of the Situation Model: Comprehending the Situation

14.2.1.2 Selection of Actions and the Control of Behaviour

14.3 Errors and the Comprehension Based-Model of Situation Awareness

14.4 Situation Awareness and In-Vehicle Information System Tasks

14.4.1 A Measurement Procedure: Context-Dependent Choice Reaction Task

14.4.2 Evaluation of the Context-Dependent Choice Reaction Task

14.5 Conclusions

References

15 Driver Error and Crashes

15.1 Slips, Lapses and Mistakes

15.2 Errors and Violations

15.3 The Manchester Driver Behaviour Questionnaire

15.4 The DBQ and Road Traffic Accidents

15.5 Aggressive Violations

15.6 Anger-Provoking Situations

15.7 Conclusions

References

VI Control Theory Models of Driver Behaviour

16 Control Theory Models of the Driver

16.1 Introduction

16.2 Modelling Human Controlling Behaviour

16.2.1 The Tustin-Model: Linear Part + Remnant

16.2.2 Laboratory Research, Stochastic Input, Quasi-Linear Model

16.2.3 A Holistic Approach: The Crossover Model

16.2.4 Nonlinear Approaches: Improved Reproduction of Measured Behaviour

16.3 Driver Models for Vehicle Design

16.4 Summary and Future Prospects

References

17 Review of Control Theory Models for Directional and Speed Control

17.1 Introduction

17.2 Basic Crossover Model of the Human Operator

17.3 Model for Driver Steering Control

17.3.1 Equivalent Single-Loop System for Steering Control

17.4 Model for Speed Control with Accelerator Pedal

17.5 Experimental Data

17.5.1 Driving Simulator Measurements

17.5.1.1 Steering Control

17.5.1.2 Speed Control

17.5.2 Actual Vehicle Measurements

17.6 Example Directional Control Application

17.7 Discussion

References

VII Simulation of Driver Behaviour

18 Cognitive Modelling and Computational Simulation of Drivers Mental Activities

18.1 Introduction: A Brief Historical Overview on Driver Modelling

18.2 COSMODRIVE Model

18.2.1 Cognitive Architecture ofCOSMODRIVE

18.2.2 The Tactical Module

18.2.2.1 Driving Frames: A Framework for Modelling Mental Models

18.2.2 .2 Architecture of the Tactical Module

18.2.2.3 The Blackboards of the Tactical Module

18.2.2.4 The Knowledge Bases (KB) of the Tactical Module

18.2.2.5 The Cognitive Processes of the Tactical Module

18.2.2.5.1 Categorisation

18.2.2.5.2 The Place Recognition Process

18.2.2.5.3 The Tactical Representations Generator Process

18.2.2.5.4 The Anticipation Process

18.2.2.5.5 The Decision Process

18.3 Methodology to Study Driver's Situation Awareness

18.3.1 Main Hypothesis

18.3.2 Methodology

18.3.3 Main Results

18.3.4 Discussion and Conclusion Concerning Experimental Study of Drivers Situation Awareness

18.4 Some Experimental Results Simulation with Cosmodrive

18.5 Conclusion and Perspectives: From Behaviours to Mental Model

References

19 Simple Simulation of Driver Performance for Prediction and Design Analysis

19.1 Introduction

19.1.1 Modelling Human Behaviour in Modern Technology

19.1.2 Modelling Drivers in the Automotive Context

19.1.3 Use and Applications ofDriver Models

19.1.4 Content ofthe Paper

19.2 Simple Simulation of Driver Behaviour

19.2.1 Paradigm of Reference

19.2.2 Simulation Approach for Normative Behaviour

19.2.2.1 Task Analysis

19.2.2.2 Dynamic Logical Simulation of Tasks

19.2.3 Algorithms for Cognition, Behavioural Adaptation and Errors

19.2.3.1 Normative Driver Behaviour

19.2.3.2 Descriptive Driver Behaviour

19.2.3.3 Parameters and Measurable Variables

19.2.3.3.1 Task Demand

19.2.3.3.2 Driver State

19.2.3.3.3 Situation Awareness

19.2.3.4 Intentions, Decision Making and Human Error

19.2.3.4.1 Intentions and Decision Making

19.2.3.4.2 Error Generation

19.2.4 Simulation of Control Actions

19.2.4.1 Normal Driving

19.2.4.2 Error in Control Actions

19.3 Sample Cases of Predictive DVE Interactions

19.3.1 Case 1

19.3.2 Case 2

19.4 Conclusions

References

VIII Simulation of Traffic and Real Situations

20 Real-Time Traffic and Environment Risk Estimation for the Optimisation of Human-Machine Interaction

20.1 Introduction

20.2 The AWAKE Use Case - Adaptation of a Driver Hypovigilance Warning System

20.2.1 AWAKE System Overview

20.2.2 Traffic Risk Estimation in AWAKE System

20.2.3 The Scenario-Assessment Unit

20.2.4 The Warning Strategies Unit

20.2.5 The Risk-Level Assessment Unit

20.3 The AIDE Use Case - Optimisation of the In-Vehicle Human-Machine Interaction

20.3.1 Overview

20.3.2 Architecture

20.3.2.1 Relevance to the AIDE Use Cases

20.3.2.2 Description of Environment

20.3.3 Algorithm for Risk Assessment

20.3.3.1 Rule-Based System Employed for TERA Algorithms

20.3.3.2 Main Traffic Risk Condition Detection

20.3.3.2.1 Risk of Frontal/(Lateral) Collision

20.3.3.2.2 Criteria of Assigning the Level of Risk

20.3.3.2.3 Risk of Lane/Road Departure

20.3.3.2.4 Risk of Approaching a Dangerous Curve Too Fast

20.3.4 Algorithmfor Estimating the Intention of the Driver

20.3.5 TERA Implementation

20.4 Conclusions

References

21 Present and Future of Simulation Traffic Models

21.1 Introduction

21.2 Traffic Simulator

21.2.1 General Overview: A Survey of Road Traffic Simulations

21.2.2 Types of Simulator

21.2.3 Case Studies of Traffic Simulator

21.2.4 Vehicle Model Properties

21.2.4.1 Perception Topics

21.2.4.2 Cognition Topics

21.2.4.3 Actuation/Control Topics

21.2.4.4 Implementation of Vehicle Model

21.2.5 Two Examples of Applications with Traffic Simulator

21.2.5.1 The University of Michigan Microscopic Traffic Simulator

21.2.5.2 The MECTRON-HMI Group at University of Modena and Reggio Emilia Driving Simulator used in Human factors and Human-Machine Interfaces Studies.

21.2.6 Integration of Driver, Vehicle and Environment in a Closed-Loop System: The AIDE Project

21.2.6.1 General DVE Architecture

21.2.6.2 Time Frame for DVE Model

21.3 Conclusions and Further Steps

21.3.1 Towards a Multi-Agent Approach

21.3.2 New Developments and Prospective

21.3.3 Open Points and Future Steps

References

Index