Han | Generative AI for Connected and Autonomous Vehicles | Buch | 978-1-394-42668-3 | www.sack.de

Buch, Englisch, 400 Seiten, Format (B × H): 185 mm x 263 mm, Gewicht: 896 g

Han

Generative AI for Connected and Autonomous Vehicles


1. Auflage 2026
ISBN: 978-1-394-42668-3
Verlag: John Wiley & Sons Inc

Buch, Englisch, 400 Seiten, Format (B × H): 185 mm x 263 mm, Gewicht: 896 g

ISBN: 978-1-394-42668-3
Verlag: John Wiley & Sons Inc


Harness generative AI for connected and autonomous vehicle systems

As generative AI reshapes intelligent transportation, engineers and researchers need rigorous frameworks for deploying these technologies in safety-critical systems. This book offers a comprehensive treatment of the models, protocols, and adaptation techniques required to bring large language models, vision-language models, diffusion models, and agentic systems into modern CAV architectures. The book progresses from generative foundations through multimodal representation, language-based interaction, agentic coordination, communication protocols, model adaptation, and validation and safety. Mathematical, probabilistic, and reinforcement-learning prerequisites are consolidated into dedicated appendices, allowing the main chapters to focus on system design and CAV-specific applications.

Written by an award-winning researcher with more than 100 patents across connected vehicles and AI, readers will also find:

- Practical treatment of prompt engineering, retrieval-augmented generation, function calling, and tool use for grounding GenAI in vehicular context
- Chapters on agent architectures, multi-agent systems, orchestration frameworks, and MCP and A2A protocols for collaborative CAV intelligence
- Adaptation and deployment techniques including supervised fine-tuning, reinforcement learning fine-tuning, and knowledge distillation for resource-constrained edge platforms
- Dedicated chapters on simulation platforms, evaluation methodologies, and ethical and safety considerations for deploying generative AI in autonomous vehicles
- Supported by instructor resources and additional supporting materials

Automotive engineers, AI researchers, and graduate students at the intersection of machine learning and intelligent transportation will find this book indispwensable. Readers from OEMs, mobility startups, or academic labs gain the foundations and practices needed to design, adapt, and validate generative AI in connected mobility.

Han Generative AI for Connected and Autonomous Vehicles jetzt bestellen!

Autoren/Hrsg.


Weitere Infos & Material


About the Author xxi
Preface xxiii
Acknowledgments xxv
How to Read This Book xxvii
Acronyms xxxi
About the Companion Website xxxv

Part I Foundations of Generative Artificial Intelligence 1

1 Introduction to Generative AI for CAV Systems 3
1.1 Introduction: The Promise of Generative AI in CAVs 3
1.2 The Paradigm Shift: Generative vs. Discriminative AI 3
1.3 Transformative Capabilities in CAVs 6
1.4 Generative Model Families 7
1.5 The Generative AI Ecosystem 11
1.6 Summary and Roadmap 15

2 Large Language Models: Architecture and Operation 19
2.1 Introduction 19
2.2 Large Language Models 20
2.3 The Rise of Transformers in Language Modeling 23
2.4 The LLM Generation Pipeline 24
2.5 Stage 1: Tokenization and Embeddings 26
2.6 Stage 2: Positional Encoding 27
2.7 Stage 3: Attention and Contextualization 29
2.8 Stage 4: Refinement and Stacking 32
2.9 Stage 5: Output and Sampling 34
2.10 Model Limitations and Comparative Landscape 36
2.11 Summary 39

Part II Multimodal Representation and World Grounding 43

3 Diffusion and Flow Matching 45
3.1 Introduction: Why Diffusion and Flow Matching for CAV Systems 45
3.2 Conceptual Foundations 46
3.3 Mathematical Foundations 47
3.4 Architectural Choices in Diffusion and Transport Models 51
3.5 Applications in CAV Systems 53
3.6 Limitations and Considerations 54
3.7 Summary 55

4 GANs, VAEs, and Autoregressive Models 57
4.1 Introduction 57
4.2 Generative Adversarial Networks 58
4.3 Variational Autoencoders 62
4.4 Autoregressive Models for Sequential Prediction 68
4.5 Comparative Analysis 72
4.6 Summary 72

5 Vision-language Models for Scene Understanding 75
5.1 Introduction 75
5.2 Architectural Principles of VLMs 76
5.3 Stage 1: Visual Feature Extraction 78
5.4 Stage 2: Vision-language Adaptation 80
5.5 Stage 3: Integration into the Language Model 82
5.6 Training Methodologies for VLMs 84
5.7 Deployment in Connected and Autonomous Vehicles 89
5.8 Case Study: Semantic Filtering of V2X Messages 90
5.9 Summary 92

6 Video Large Language Models for Temporal Reasoning in Driving Environments 95
6.1 Introduction 95
6.2 Fundamental Design Principles 96
6.3 Stage 1: Spatio-temporal Encoding 97
6.4 Stage 2: Spatio-temporal Fusion 99
6.5 Stage 3: Integration with Language Models 100Contents xi
6.6 Token Compression Techniques 103
6.7 Representative Architectures 105
6.8 Applications in Connected and Autonomous Vehicles 107
6.9 Summary 108

Part III Language-based Interaction, Memory, and Tool Use 111

7 Prompt Engineering 113
7.1 Introduction and Motivation 113
7.2 Foundations of Prompt Engineering 114
7.3 Design Principles for Effective Prompting 116
7.4 Prompt Engineering in Practice 119
7.5 Prompt Fragility and Safety Considerations in CAV Systems 123
7.6 Summary 124

8 Retrieval-augmented Generation for Context Grounding 127
8.1 Why RAG Matters in Connected and Autonomous Vehicles 127
8.2 Architectural Foundations of RAG 129
8.3 RAG vs. Rule-based Information Systems in CAVs 130
8.4 Mathematical Modeling of RAG 132
8.5 Practical RAG Implementation for CAV Systems 133
8.6 Real-time Feasibility and System Constraints 135
8.7 Applied Use Cases of RAG in CAV Systems 136
8.8 Summary 137

9 Function Calling and Tool Use 139
9.1 Introduction: Why Connected and Autonomous Vehicles Need Function Calling 139
9.2 System Architecture of Function Calling 140
9.3 Theoretical Foundations of Function Calling 143
9.4 Function Calling in Practice 145
9.5 Function Schema Library and Design Principles 148
9.6 Summary 152

Part IV Agentic Intelligence and Coordination 155

10 GenAI Agent Architectures 157
10.1 Introduction 157
10.2 Classical Agent Foundations and Architecture 158
10.3 GenAI Agent Mechanisms 160
10.4 Mathematical Formalization 163
10.5 ReAct Framework 165
10.6 Individual Agent Applications in CAVs 166
10.7 Limitations and Challenges of GenAI Agents 167
10.8 Summary 169

11 GenAI Multi-agent Systems 171
11.1 Introduction 171
11.2 Multi-agent Systems Foundations and Architecture 172
11.3 GenAI Coordination Mechanisms 174xiv Contents
11.4 Mathematical Formalization 177
11.5 GenAI Applications in CAVs 179
11.6 Implementation Considerations and Practical Challenges 180
11.7 Summary 183

12 Orchestration Frameworks for Agentic AI 185
12.1 Introduction 185
12.2 Graph-based Orchestration with LangGraph 186
12.3 Dialogue-based Orchestration with AutoGen 191
12.4 Framework Comparison and Selection 195
12.5 Summary 196

Part V Communication Protocols and Collective Intelligence 199

13 Communication Protocols for Agentic AI 201
13.1 Introduction 201
13.2 Protocol Classification Framework 202
13.3 Protocol Evolution and Development 205
13.4 Protocol Examples and CAV Applications 206
13.5 Summary 207

14 Model Context Protocol for Agentic Systems 209
14.1 Introduction 209
14.2 Context Provision in CAV Systems 209
14.3 MCP Architecture 211
14.4 Server Features 215
14.5 Client Features 218
14.6 Mathematical Foundations 220
14.7 Practical Implementation 221
14.8 MCP in CAV Systems 223
14.9 Summary 225

15 Agent-to-agent Protocols for Collaborative Intelligence 227
15.1 Introduction 227
15.2 A2A Architecture 227
15.3 Mathematical Foundations for Agent Collaboration 232
15.4 Security and Trust 235
15.5 A2A in CAV Systems 236
15.6 Limitations and Challenges 238
15.7 Summary 239

16 Integrated CAV Agent Architectures with MCP and A2A 241
16.1 Introduction 241
16.2 Protocol Comparison and Decision Framework 241
16.3 System Architecture and Mathematical Framework 243
16.4 Implementation and Execution Patterns 245
16.5 Summary 249

Part VI Adaptation and Deployment 251

17 Supervised Fine-tuning 253
17.1 The Necessity of Fine-tuning in CAV Applications 253
17.2 From Foundation Models to Specialization 254
17.3 Architectural Perspective on Fine-tuning 255
17.4 Mathematical Formulation of Fine-tuning 256
17.5 Practical Fine-tuning 258
17.6 Parameter-efficient Fine-tuning Methods 260
17.7 Use Cases and Deployment Modes 265
17.8 Summary 268

18 Reinforcement Learning Fine-tuning 271
18.1 Introduction 271
18.2 Mathematical Foundation 273
18.3 Policy Optimization Methods in Detail 277
18.4 RLHF Methodology 284
18.5 Summary 286

19 Knowledge Distillation for Edge Deployment 289
19.1 Introduction: The Deployment Challenge 289
19.2 The Teacher–Student Paradigm 289
19.3 Core Knowledge Distillation Mechanisms 290
19.4 Advanced Distillation Techniques 294
19.5 Deployment Considerations for CAV Applications 296
19.6 Summary 298

Part VII Validation and Safety 299

20 Simulation Platforms for GenAI-driven CAVs 301
20.1 Introduction: Why Simulation Is Indispensable for GenAI-CAVs 301
20.2 Simulation as a System-level Substrate for GenAI-CAVs 302
20.3 Scenario Generation and Coverage Expansion 303
20.4 Simulation Platforms for GenAI-CAV Development 304
20.5 Scaling Simulation: From Single-agent to Multi-agent Systems 307
20.6 Simulation-evaluation Boundary 308
20.7 Summary 308

21 Evaluation of Generative AI Systems in CAVs 311
21.1 Introduction: The Role of Evaluation in GenAI-driven CAVs 311
21.2 Evaluation Dimensions for GenAI Systems in CAV Applications 312
21.3 Fundamental Evaluation Metrics and Methodologies 313
21.4 Evaluation of Vision-language Models for CAV Applications 324
21.5 Evaluation of Retrieval-augmented Generation Systems 326
21.6 Multi-turn Dialogue and Conversational Agents 329
21.7 Summary 330

22 Ethical and Safety Considerations 333
22.1 Introduction: The Ethical Stakes of GenAI in CAVs 333
22.2 Foundational Ethical Principles and Challenges 334
22.3 Case Studies: Ethical Challenges in Practice 342
22.4 Regulatory and Safety Frameworks 343
22.5 Ethical Engineering Methods 345
22.6 Broader Environmental and Societal Considerations 348
22.7 Summary 349

References 350
Index 353


KYUNGTAE HAN, PhD, is a Senior Principal Researcher at Toyota Motor North America and was previously a Senior Research Scientist at Intel Labs. With over 90 publications and more than 100 patents spanning generative AI, connected and automated vehicles, cyber-physical systems, and edge computing, his work has earned the SAE Vincent Bendix Award, IEEE Best Application Award, and Intel Achievement Award.



Ihre Fragen, Wünsche oder Anmerkungen
Vorname*
Nachname*
Ihre E-Mail-Adresse*
Kundennr.
Ihre Nachricht*
Lediglich mit * gekennzeichnete Felder sind Pflichtfelder.
Wenn Sie die im Kontaktformular eingegebenen Daten durch Klick auf den nachfolgenden Button übersenden, erklären Sie sich damit einverstanden, dass wir Ihr Angaben für die Beantwortung Ihrer Anfrage verwenden. Selbstverständlich werden Ihre Daten vertraulich behandelt und nicht an Dritte weitergegeben. Sie können der Verwendung Ihrer Daten jederzeit widersprechen. Das Datenhandling bei Sack Fachmedien erklären wir Ihnen in unserer Datenschutzerklärung.