Villagra / Jiménez Decision-Making Techniques for Autonomous Vehicles

1. Auflage 2023
ISBN: 978-0-323-98549-9
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, 424 Seiten

Decision-Making Techniques for Autonomous Vehicles provides a general overview of control and decision-making tools that could be used in autonomous vehicles. Motion prediction and planning tools are presented, along with the use of machine learning and adaptability to improve performance of algorithms in real scenarios. The book then examines how driver monitoring and behavior analysis are used produce comprehensive and predictable reactions in automated vehicles. The book ultimately covers regulatory and ethical issues to consider for implementing correct and robust decision-making. This book is for researchers as well as Masters and PhD students working with autonomous vehicles and decision algorithms. - Provides a complete overview of decision-making and control techniques for autonomous vehicles - Includes technical, physical, and mathematical explanations to provide knowledge for implementation of tools - Features machine learning to improve performance of decision-making algorithms - Shows how regulations and ethics influence the development and implementation of these algorithms in real scenarios

Autoren/Hrsg.

Villagra, Jorge

Jiménez, Felipe

Inhaltsverzeichnis

1;Front Cover;1
2;Decision-making Techniques for Autonomous Vehicles;4
3;Copyright;5
4;Contents;6
5;Contributors;12
6;About the editors;14
7;Chapter 1: Overview;18
7.1;1.1. Introduction;18
7.2;1.2. Decision-making, automation levels, and operational design domains;20
7.3;1.3. Scope of the book;21
7.4;1.4. Book structure overview;24
7.5;References;30
8;Part I: Embedded decision components;32
9;Chapter 2: Embodied decision architectures;34
9.1;2.1. Introduction;34
9.2;2.2. Embodiment and cognitive capabilities;34
9.3;2.3. Cognitive architectures and biological plausible human behavioral models;37
9.4;2.4. Decision architectures for autonomous driving;40
9.4.1;2.4.1. Examples of subsumption architectures;41
9.4.2;2.4.2. An ADAS-oriented behavior architecture;44
9.4.3;2.4.3. Examples of cognition-inspired architectures;45
9.4.4;2.4.4. Safety-oriented architectures;47
9.4.5;2.4.5. Shared control architectures;49
9.5;2.5. Common functional blocks;51
9.6;References;53
10;Chapter 3: Behavior planning;56
10.1;3.1. Introduction;56
10.2;3.2. Problem description;57
10.3;3.3. Automata and Markov processes;60
10.4;3.4. Basic decision theory;61
10.5;3.5. Sequential decision-making;65
10.6;3.6. Applications in automated vehicles;67
10.6.1;3.6.1. Rule-based planning;68
10.6.2;3.6.2. Reactive planning;69
10.6.3;3.6.3. Interaction-aware planning;70
10.6.4;3.6.4. Game theory for behavior planning;71
10.6.5;3.6.5. AI-enabled behavior planning;72
10.7;References;74
11;Chapter 4: Motion prediction and risk assessment;78
11.1;4.1. Introduction;78
11.1.1;4.1.1. Problem formalization;79
11.2;4.2. Driver trait estimation;81
11.2.1;4.2.1. Scope;81
11.2.2;4.2.2. Representation;81
11.2.3;4.2.3. Inference approaches;82
11.3;4.3. Intention estimation;82
11.3.1;4.3.1. Scope;83
11.3.2;4.3.2. Representation;84
11.3.3;4.3.3. Inference approaches;84
11.3.3.1;4.3.3.1. Recursive estimation algorithms;84
11.3.3.2;4.3.3.2. Bayesian approaches;84
11.3.3.3;4.3.3.3. Game-theoretic approaches;85
11.3.3.4;4.3.3.4. Learning-based approaches;85
11.4;4.4. Motion prediction;91
11.4.1;4.4.1. Scope;91
11.4.2;4.4.2. Representation;92
11.4.2.1;4.4.2.1. Agent-level representation;94
11.4.2.2;4.4.2.2. Scenario level representation;95
11.4.3;4.4.3. Modeling approaches;96
11.4.3.1;4.4.3.1. Physics-based;96
11.4.3.1.1;Single-model approaches;96
11.4.3.1.2;Multimodel approaches;98
11.4.3.2;4.4.3.2. Learning-based approaches;99
11.4.3.2.1;Sequential;99
11.4.3.2.2;Nonsequential;101
11.4.3.3;4.4.3.3. Planning-based;102
11.4.3.3.1;Forward planning;102
11.4.3.3.2;Inverse planning;102
11.4.4;4.4.4. Situational awareness considerations;103
11.4.4.1;4.4.4.1. Unaware;103
11.4.4.2;4.4.4.2. Intention-aware;103
11.4.4.3;4.4.4.3. Scene awareness;104
11.4.4.4;4.4.4.4. Map aware;104
11.4.5;4.4.5. Metrics;105
11.4.5.1;4.4.5.1. Geometric accuracy metrics;105
11.4.5.2;4.4.5.2. Probabilistic accuracy metrics;106
11.5;4.5. Risk assessment;106
11.5.1;4.5.1. Scope;106
11.5.2;4.5.2. Representation;107
11.5.3;4.5.3. Inference strategies;108
11.5.3.1;4.5.3.1. Risk based on future trajectories;108
11.5.3.1.1;Binary collision prediction;108
11.5.3.1.2;Collision risk based on indicators;109
11.5.3.1.3;Probabilistic collision prediction;110
11.5.4;4.5.4. Risk based on unexpected behavior;111
11.5.4.1;4.5.4.1. Detecting unusual events;111
11.5.4.2;4.5.4.2. Detecting conflicting maneuvers;112
11.6;References;112
12;Chapter 5: Motion search space;120
12.1;5.1. Introduction;120
12.2;5.2. Graph-based techniques;121
12.3;5.3. Geometric methods;122
12.3.1;5.3.1. Nonobstacle-based techniques;124
12.3.2;5.3.2. Obstacle-based techniques;125
12.4;5.4. Sampling-based methods;126
12.5;5.5. Driving corridors;130
12.6;References;131
13;Chapter 6: Motion planning;134
13.1;6.1. Problem definition;134
13.1.1;6.1.1. Brief taxonomy of motion planners;136
13.2;6.2. Geometric methods;138
13.2.1;6.2.1. Point-free template-based geometric strategies;138
13.2.2;6.2.2. Point-based template-based curves;140
13.3;6.3. Variational and optimal methods;142
13.3.1;6.3.1. MPC architecture;144
13.3.2;6.3.2. Optimization techniques;145
13.3.3;6.3.3. Local nonconvex optimization;146
13.3.4;6.3.4. Global nonconvex optimization;147
13.3.5;6.3.5. Relevant uses cases;152
13.3.5.1;6.3.5.1. Strategies considering comfort and safety;152
13.3.5.2;6.3.5.2. Obstacle avoidance and overtaking maneuvers;153
13.4;6.4. Sampling-based methods;155
13.4.1;6.4.1. General formulation of the deterministic problem;155
13.4.1.1;6.4.1.1. Metric state and measure function;156
13.4.1.2;6.4.1.2. Sampling strategy;156
13.4.1.3;6.4.1.3. Collision detection and path segment validation;157
13.4.2;6.4.2. Multiple-query and single-query methods;158
13.4.2.1;6.4.2.1. Single-query methods;158
13.4.2.2;6.4.2.2. Multiple-query methods;158
13.4.2.3;6.4.2.3. Deterministic approaches in autonomous driving;160
13.4.3;6.4.3. General formulation of the probabilistic problem;161
13.4.3.1;6.4.3.1. Probabilistic approaches in autonomous driving;161
13.4.4;6.4.4. Sampling-based methods with constraints;162
13.5;6.5. Graph-search methods;163
13.6;6.6. Cognition-inspired approaches;165
13.6.1;6.6.1. Evolutionary computation;166
13.6.2;6.6.2. Fuzzy logic and neural networks;167
13.7;6.7. Biomimetic methods;167
13.7.1;6.7.1. Artificial potential fields;167
13.7.2;6.7.2. Elastic bands;171
13.8;6.8. From vehicle following/CACC to standalone speed planning;172
13.9;6.9. Separated speed planning;174
13.10;6.10. Joint path and speed optimization-based planning;177
13.11;References;178
14;Chapter 7: End-to-end architectures;186
14.1;7.1. End-to-end approaches;186
14.1.1;7.1.1. Introduction;186
14.2;7.2. End-to-end approaches based on deep learning;188
14.2.1;7.2.1. Classification of end-to-end architectures for autonomous driving;190
14.2.1.1;7.2.1.1. SiD-E2E architecture;190
14.2.1.2;7.2.1.2. MiD-E2E architecture;191
14.2.1.3;7.2.1.3. SeD-E2E architecture;192
14.2.2;7.2.2. Transfer learning vs ad-hoc solutions;194
14.2.2.1;7.2.2.1. Transfer learning;194
14.2.2.2;7.2.2.2. Ad-hoc solution;195
14.2.3;7.2.3. Datasets for modeling end-to-end solutions;196
14.2.4;7.2.4. Reinforcement learning techniques;199
14.3;7.3. Expert systems;203
14.4;7.4. Future perspectives;205
14.5;References;206
15;Chapter 8: Interplay between decision and control;210
15.1;8.1. Introduction;210
15.2;8.2. Stabilization principles;211
15.2.1;8.2.1. Problem definition;211
15.2.2;8.2.2. Stabilization requirements for automated driving;212
15.2.3;8.2.3. Longitudinal control;214
15.2.4;8.2.4. Lateral control;216
15.3;8.3. Upstream vs downstream control architectures;218
15.4;8.4. Interaction models between planning and control;222
15.4.1;8.4.1. Ethical considerations;222
15.4.2;8.4.2. Integrated vs decoupled planning and control;224
15.5;8.5. Fail-operational considerations;226
15.6;References;227
16;Part II: Infrastructure-oriented decision-making;232
17;Chapter 9: Traffic data analysis and route planning;234
17.1;9.1. Introduction;234
17.2;9.2. Off-board decision-making: From the traveling salesman problem to the vehicle routing problem;235
17.3;9.3. On the relevance of traffic data and exogenous information for predictive route planning;242
17.3.1;9.3.1. Considering traffic forecasts;243
17.3.2;9.3.2. Short-term traffic forecasting;243
17.3.3;9.3.3. Long-term traffic forecasting;248
17.4;9.4. Challenges and research directions in the confluence between route planning and traffic data analysis;250
17.4.1;9.4.1. From route optimization toward learning to route;250
17.4.2;9.4.2. Causal agent-based traffic models and route planning;252
17.4.3;9.4.3. Knowledge transfer for route optimization;253
17.4.4;9.4.4. Toward explainable and trustworthy route planning;254
17.5;References;255
18;Chapter 10: Cooperative driving;262
18.1;10.1. Introduction to cooperative, connected, and automated driving (3P);262
18.1.1;10.1.1. Solution: CCAD;264
18.1.1.1;10.1.1.1. Onboard decision-making;264
18.1.1.2;10.1.1.2. Decision-making in infrastructure;265
18.2;10.2. Communication technologies;265
18.2.1;10.2.1. Vehicle-to-everything (V2X);266
18.2.2;10.2.2. DSRC (IEEE 802.11p, ETSI ITS-G5) (V2X standards overview);267
18.2.2.1;10.2.2.1. IEEE 802.11p;268
18.2.3;10.2.3. Cellular V2X;268
18.2.4;10.2.4. Security;269
18.2.4.1;10.2.4.1. SerIoT project;270
18.3;10.3. Connected services;271
18.4;10.4. Adaptation of decision-making mechanisms to support V2X;273
18.4.1;10.4.1. Connected and automated scenario for cyber-attacks (SerIoT project);273
18.4.1.1;10.4.1.1. Smart intersection in normal situation;274
18.4.1.2;10.4.1.2. Fleet management in normal situation;275
18.4.2;10.4.2. Platoon maneuver;276
18.4.3;10.4.3. Roundabout merging scenarios maneuvers;277
18.4.4;10.4.4. Conclusions;277
18.5;References;278
19;Chapter 11: Infrastructure impact;280
19.1;11.1. The role of the physical infrastructure: From evidence to guidelines;280
19.1.1;11.1.1. Impact of road infrastructure on automated driving;280
19.1.1.1;11.1.1.1. Road typology;280
19.1.1.2;11.1.1.2. Geometry;281
19.1.1.3;11.1.1.3. Road markings;282
19.1.1.4;11.1.1.4. Traffic signs;284
19.1.1.5;11.1.1.5. Junctions;284
19.1.1.6;11.1.1.6. Pavement condition;285
19.1.1.7;11.1.1.7. Road environment;285
19.1.1.8;11.1.1.8. Environmental conditions;286
19.1.1.9;11.1.1.9. Road works and temporary emergency signage;286
19.1.1.10;11.1.1.10. Speed;286
19.2;11.2. Information required from infrastructure to enable different ad levels;287
19.2.1;11.2.1. Specifications of road infrastructure for different AD levels;287
19.2.1.1;11.2.1.1. Level of service for automated driving (LOSAD);287
19.2.1.2;11.2.1.2. Infrastructure support levels for automated driving (ISAD);289
19.2.1.3;11.2.1.3. Smart road levels (SRL);289
19.2.1.4;11.2.1.4. Factors related to physical infrastructure;291
19.2.1.5;11.2.1.5. Information management providers;292
19.2.2;11.2.2. Minimal risk condition;293
19.3;References;294
20;Part III: User influence;298
21;Chapter 12: Driver behavior;300
21.1;12.1. A human-centered perspective in driving automation;300
21.2;12.2. Human-driver assessment from the perspective of HAI models in automated driving;310
21.2.1;12.2.1. Definition and assessment of mental workload (MWL);311
21.2.2;12.2.2. Reduced situation awareness (SA);314
21.2.3;12.2.3. Complacency or overtrust;316
21.2.4;12.2.4. Skill degradation and the loss of sense of authority;317
21.2.5;12.2.5. Control transition between automation and human driver in automated vehicles;317
21.3;12.3. Passengers in autonomous vehicles;322
21.3.1;12.3.1. The changing role of the passenger in autonomous vehicles;322
21.3.2;12.3.2. The acceptance of autonomous vehicles;323
21.3.2.1;12.3.2.1. Personal expectations;324
21.3.2.2;12.3.2.2. Perceived attributes;325
21.3.2.3;12.3.2.3. Personality factors;325
21.3.3;12.3.3. Emotional state of the passenger in autonomous vehicles;326
21.3.4;12.3.4. Driver attributes and external factors that affect the passenger state;334
21.3.4.1;12.3.4.1. External factors;334
21.3.4.2;12.3.4.2. Ego vehicle factors;335
21.3.4.3;12.3.4.3. Internal factors;335
21.3.5;12.3.5. Human machine interface for passengers in autonomous vehicles;336
21.3.6;12.3.6. Ride, ambient comfort, well-being, and other services;337
21.4;References;341
22;Chapter 13: Human-machine interaction;350
22.1;13.1. Introduction;350
22.2;13.2. Human-Machine Cooperation and metaphors used in shared control;352
22.3;13.3. Shared control approaches;353
22.3.1;13.3.1. Definitions;353
22.3.2;13.3.2. Frameworks;355
22.3.3;13.3.3. Algorithms;356
22.4;13.4. A recent human-machine interaction framework;357
22.5;13.5. Traded control mechanisms;360
22.5.1;13.5.1. Suitability of traded control;360
22.5.2;13.5.2. Traded control (de)activation principles;363
22.6;References;366
23;Part IV: Deployment issues;370
24;Chapter 14: Algorithms validation;372
24.1;14.1. Introduction;372
24.2;14.2. Validation methodology;373
24.2.1;14.2.1. Testing process;374
24.2.2;14.2.2. Main techniques for ADF validation;375
24.2.3;14.2.3. Datasets;377
24.3;14.3. Simulation systems;378
24.3.1;14.3.1. Human-in-the-Loop simulation (HITL);380
24.3.1.1;14.3.1.1. Human-in-the-Loop AI;381
24.3.1.2;14.3.1.2. Human-in-the-Loop driving simulation;382
24.3.2;14.3.2. Vehicle-in-the-loop simulation;383
24.3.2.1;14.3.2.1. Definition;383
24.3.2.2;14.3.2.2. Traffic simulation: Scene simulation;385
24.4;14.4. Standards for safety assurance;386
24.5;References;387
25;Chapter 15: Legal and social aspects;392
25.1;15.1. Introduction;392
25.2;15.2. Regulation;392
25.2.1;15.2.1. Introduction;393
25.2.2;15.2.2. International governance;394
25.2.3;15.2.3. The Vienna convention on road traffic;395
25.2.3.1;15.2.3.1. Amendments to the Vienna convention on road traffic;396
25.2.4;15.2.4. State-of-the-art on the European regulations on autonomous vehicles;397
25.2.4.1;15.2.4.1. Context of European regulations;397
25.2.4.2;15.2.4.2. Comparative analysis of international regulations;398
25.3;15.3. Ethics;398
25.3.1;15.3.1. Ethical problem of autonomous driving;398
25.3.2;15.3.2. Approaches to face the ethical problem;402
25.3.3;15.3.3. Conclusions;407
25.4;15.4. User acceptance;407
25.4.1;15.4.1. Introduction;407
25.4.2;15.4.2. Perceived safety;409
25.4.3;15.4.3. Trust;410
25.4.4;15.4.4. Demographic factors;411
25.4.5;15.4.5. Psychological factors;412
25.5;References;413
26;Index;418
27;Back Cover;426