Automated Driving (eBook)

Daniel Watzenig was born in Austria. He received his Master degree in electrical engineering and the doctoral degree in technical science from Graz University of Technology, Graz, Austria, in 2002 and 2006, respectively. He is currently divisional director of the automotive electronics and embedded software department of the Virtual Vehicle Research Center in Graz. Since 2009 he is an Associate Professor at the Institute of Electrical Measurement and Measurement Signal Processing, Graz University of Technology, Austria. He is author or co-author of over 120 peer-reviewed papers, book chapters, patents, and articles. His research interests focus on sensor fusion and signal processing, automotive control systems, uncertainty estimation, and robust optimization methods. In 2005 he was a visiting researcher at the University of Auckland, New Zealand, working on multi-sensor arrays and statistical signal processing. In 2011 he was visiting researcher and guest lecturer at the Federal University of Rio de Janeiro. He is IEEE Senior Member of the IEEE Control Systems, Signal Processing and Instrumentation & Measurement Societies. Furthermore, he is Vice President and member of the steering board of the EU ARTEMIS Industrial Association.   Martin Horn was born in Austria. He received his Master degree in electrical engineering and the doctoral degree in technical science from Graz University of Technology, Graz, Austria, in 1992 and 1998, respectively. From 2003 to 2008 he was is an Associate Professor at the Institute of Automation and Control at Graz University of Technology. From 2008 to 2014 he was full professor for Control and Mechatronic Systems at Klagenfurt University.  Since 2014 he is full professor for Automation and Control at Graz University of Technology. He is author or co-author of around 100 peer-reviewed papers, book chapters, patents, and articles. His research interests focus on automotive control systems, robust control, variable structure systems and robust observer design.  

Foreword 5
Preface 7
Contents 9
Contributors 13
Part I Introduction 18
1 Introduction to Automated Driving 19
1.1 Introduction 19
1.2 Levels of Automation 20
1.2.1 Scenarios and Impact of Automation Levels 2–5 According to [7] 21
1.3 Building Blocks for Automated Driving: Key Technologies 22
1.4 Enabling Automated Driving: The Research Challenges 24
1.4.1 Demonstrating Safety, Reliability, and Robustness 26
1.4.2 Demonstrating Security and Privacy 27
1.4.3 Dependable Power Computing 28
1.4.4 Human Factor (SAE Level 3/4) 28
1.4.5 Environment Modeling and Perception 29
1.4.6 Vehicle Control and Actuation 29
1.4.7 Digital Infrastructure 30
1.5 Conclusion 31
References 31
2 Privacy and Security in Autonomous Vehicles 33
2.1 Introduction and Definition of Terminology 33
2.2 Principles of Autonomous Driving 34
2.2.1 Technological Principles 34
2.2.2 Data Principles 35
2.3 Status of What Exists Today 36
2.4 Future Expectations for Autonomous Driving 37
2.5 Building Social Trust 39
2.6 Impact on Industry 40
2.7 Next Steps 42
2.8 Conclusions 43
3 Automated Driving from the View of Technical Standards 44
3.1 Introduction 44
3.2 Standard Developing Organizations 45
3.3 Standard Developing Organizations 46
3.4 Standard Developing Organizations 47
3.4.1 Vehicle-Related Standards 48
3.4.2 Communication-Related Standards 48
3.5 Road Map of Automation and Future Standardization 53
References 54
Part II The Importance of Control for Automated Driving 56
4 Survey on Control Schemes for Automated Driving on Highways 57
4.1 Introduction 57
4.1.1 Problem Statement 58
4.1.2 Trajectory Generation 60
4.1.2.1 Polynomial Approach 61
4.1.3 Control Concepts 62
4.1.3.1 PID Control 63
4.1.3.2 Fuzzy Control 64
4.1.3.3 Neural Networks 64
4.1.3.4 Linear Quadratic Regulator 65
4.1.3.5 Feedback Linearization 65
4.1.3.6 Sliding Mode Control 65
4.1.3.7 Model Predictive Control 65
4.1.3.8 H? Control 66
4.1.4 Partitioning the Problem 67
4.2 Fuzzy Control 68
4.3 Linear State Feedback Control 69
4.4 Sliding Mode Control 69
4.5 Model Predictive Control 71
4.6 Other Concepts 72
4.7 Autonomous Vehicles 74
4.8 Comparison 75
4.8.1 Simulation Example 75
4.9 Outlook 78
References 78
5 Path Tracking for Automated Driving: A Tutorial on Control System Formulations and Ongoing Research 84
Nomenclature 84
5.1 Introduction 92
5.2 Methods Based on Geometric and Kinematic Relationships 94
5.2.1 Pure Pursuit Method 94
5.2.2 Stanley Method 94
5.2.3 Chained Controller Based on Vehicle Kinematics 95
5.3 Methods Based on Conventional Feedback Controllers and Simplified Vehicle Dynamics Models 98
5.3.1 Simple Feedback Formulations 98
5.3.2 Linear Quadratic Regulators 109
5.3.2.1 Basic Linear Quadratic Formulation 109
5.3.2.2 Linear Quadratic Regulator with Feedforward Contribution 112
5.3.2.3 Linear Quadratic Regulator with Preview 113
5.3.2.4 Frequency-Shaped Linear Quadratic Control 114
5.4 Other Control Structures for Path Tracking and Remarks 115
5.4.1 Sliding Mode Controllers 115
5.4.2 Other Control Structures 120
5.4.3 Remarks 121
5.5 Recent Advances in Path Tracking Control 123
5.5.1 Advanced Feedforward and Feedback Controllers for Limit Cornering 124
5.5.2 Model Predictive Control 131
5.6 Concluding Remarks 147
Appendix: Definitions of Invariant Sets, Minkowski Sum , and Pontryagin Difference 149
References 150
6 Vehicle Reference Lane Calculation for Autonomous Vehicle Guidance Control 154
6.1 Introduction 154
6.2 Vehicle Lane-Keeping Boundaries and Requirements 155
6.3 Vehicle Driving Situation Analysis 158
6.4 Model-Based Reference Lane Calculation Method 160
6.5 Functional System Architecture Overview 166
6.6 Example Driving Situations and Module Performance 168
6.7 Conclusion 170
References 170
Part III Advances in Environment Sensing, Sensor Fusion, and Perception 172
7 The Role of Multisensor Environmental Perception for Automated Driving 173
7.1 Introduction 173
7.2 State of Practice 175
7.2.1 Dynamic Environments 176
7.2.1.1 State Estimation by Bayesian Filtering 177
7.2.1.2 Data Association 178
7.2.1.3 Existence Estimation 181
7.2.1.4 Track Management 182
7.2.2 Static Environments by Occupancy Grid Mapping 182
7.3 Challenges in Data Fusion 183
7.3.1 Sensor Characterization 183
7.3.2 Extended Objects 185
7.3.3 Track Initialization 186
7.3.4 Asynchronous Sensors and Out-of-Sequence Processing 187
7.4 Implementation Workflows and Paradigms for Perception 188
7.4.1 Design Paradigms for Perception Software 188
7.4.2 Software Environments for Testing and Validation 191
7.5 Conclusions 193
References 193
8 Galileo-Based Advanced Driver Assistance Systems: Key Components and Development 195
8.1 Introduction 195
8.2 Test Environment Aldenhoven Testing Center and automotiveGATE 196
8.3 Galileo-Based Sensor Fusion 197
8.3.1 GNSS Characteristics 197
8.3.2 Sensor Fusion 198
8.3.3 Kalman Filter, Extended Kalman Filter 199
8.3.4 Example: Simple 2D Case 200
8.3.5 Example: 3D Case 202
8.4 Applications Examples 203
8.4.1 Application 1: Cooperative Adaptive Cruise Control 204
8.4.2 Distance Determination Between Two Vehicles 204
8.4.3 Design of the Distance Controller 206
8.4.4 Experimental Results 206
8.4.5 Application 2: Collision Avoidance System 207
8.5 Conclusion 210
References 211
9 Digital Maps for Driving Assistance Systems and Autonomous Driving 213
9.1 Introduction 213
9.2 Ontology-Based Situation Understanding 216
9.2.1 Ontologies 216
9.2.2 Situation Understanding 218
9.2.2.1 Methods for Situation Understanding 219
9.2.3 Framework for Ontology-Based Situation Understanding 221
9.2.3.1 Observations 222
9.2.3.2 World Model Principles 222
9.2.3.3 Situation Understanding 223
9.2.4 Implementation and Experimental Evaluation 229
9.2.4.1 Case Study Using Manual Data 229
9.2.4.2 Case Study Using Recorded Data 232
9.2.5 Discussion 236
9.3 Map Error Detection 237
9.3.1 Definitions 238
9.3.2 Pathology 239
9.3.2.1 Structural Faults 240
9.3.2.2 Geometric Fault 241
9.3.2.3 Attributes Faults 242
9.3.3 Page's Trend Test 243
9.3.3.1 Signal Generation 244
9.3.3.2 Formulation of the Test 245
9.3.3.3 Experimental Evaluation of Fault Detection 249
9.3.4 Discussion 253
9.4 Summary 254
References 254
10 Radar Sensors in Cars 257
10.1 Introduction 257
10.2 Forward Looking Radar (FLR) 259
10.3 Blind Spot Detection Radar 262
10.4 Early Systems and Their Results 265
10.5 Trends 268
10.6 Future Directions 270
Further Reading 273
Part IV In-Vehicle Architectures and Dependable Power Computing 274
11 System Architecture and Safety Requirements for Automated Driving 275
11.1 Toward Automated Driving 275
11.1.1 Traffic Jam Pilot 277
11.1.2 Highway Pilot 277
11.2 System Architecture 277
11.2.1 Surround Sensors 278
11.2.2 Perception 279
11.2.3 Localization 280
11.2.4 Decision Making 281
11.3 Functional Safety Concept 282
11.4 Technical Safety Concept 285
11.5 Requirements Imposed on the Onboard Network by Automated Driving Functions 286
11.5.1 Requirements for the Electric Power Supply 286
11.5.2 Requirements for the Communication Network 286
11.6 Requirement Implications 287
11.7 Safety Architecture Solutions 289
11.8 Conclusion 291
References 292
12 Advanced System-Level Design for Automated Driving 294
12.1 Motivation 294
12.2 State-of-the-Art 295
12.2.1 Overview on Current Embedded System Design 295
12.2.1.1 Current Embedded Automotive Hardware 295
12.2.1.2 Timing Analysis for Automotive Systems 296
12.2.1.3 Automotive Multitasking 297
12.2.1.4 Current Embedded Prototyping 297
12.2.2 Challenges in Hardware/Software Co-design for Automated Driving: A Case Study 299
12.2.2.1 A Vision-Based Traffic Light Detection 299
12.2.2.2 Systematic Hardware/Software Co-design 299
12.2.3 Obstacles to Efficient Realizations of Automated Driving 302
12.2.3.1 Performance Gap 302
12.2.3.2 Utilization Gap 302
12.2.3.3 Development Gap 303
12.2.3.4 Scalability Gap 303
12.3 Concepts for Efficient Future Automated Driving 303
12.3.1 Bridging the Performance Gap with Heterogeneity 304
12.3.2 Bridging the Utilization Gap 304
12.3.2.1 Runtime Resource Management 305
12.3.2.2 A Runtime Fail-Operational Mechanism 306
12.3.3 Bridging the Development Gap Using Virtual Prototypes 308
12.3.3.1 Simulation of Software Execution 308
12.3.3.2 Simulation of Hardware 309
12.3.4 Bridging the Scalability Gap for Future Platforms 310
12.4 Modern Platform-Based Automotive System Design 312
12.4.1 Tooling Framework 312
12.4.1.1 Platform Generation 313
12.4.1.2 Platform Creation 313
12.4.1.3 Platform Execution 314
12.4.2 Evaluation 314
12.4.2.1 Runtime Resource Management 314
12.4.2.2 Platform Timing Simulation 316
12.5 Conclusions 317
References 317
13 Systems Engineering and Architecting for Intelligent Autonomous Systems 321
13.1 Introduction 321
13.2 Research Method 323
13.3 Essential Terminology and Concepts 325
13.4 The Context of Machine Consciousness 326
13.5 An Architecture for Autonomous Driving 331
13.5.1 Main Architectural Components 331
13.5.2 A Reference Architecture 333
13.5.2.1 Comparison with Similar Architectures 336
13.6 Systems Engineering 338
13.7 Technical Implementation 344
13.8 Discussion 348
13.8.1 A Holistic View 349
13.8.2 The Influence of Autonomy 352
13.8.3 Concluding Remarks and Future Work 354
References 355
14 Open Dependable Power Computing Platform for Automated Driving 360
14.1 *-6pt 360
14.2 Requirements for an Open Dependable Power Computing Platform 363
14.3 Why Qualifiable Open Source Is Appropriate 366
14.4 Considerations for the Platform Architecture 368
14.5 Steps Toward an Open Dependable Computing Platform 369
14.6 Open-Source Software Development Process 371
14.7 Conclusion 372
References 373
Part V Active and Functional Safety in Automated Driving 375
15 Active Safety Towards Highly Automated Driving 376
15.1 Introduction 376
15.1.1 Motivation for Automated Driving 376
15.1.2 Development of Automated Driving Functions 378
15.1.3 Introduction of Highly Automated Driving: Motorway 378
15.1.4 Direct Safety Benefit 379
15.1.5 Indirect Safety Benefit 380
15.2 Future Development of Active Safety Systems 382
15.2.1 Required Technologies: Highly Automated Driving 382
15.2.2 Difference Between Highly Automated Driving and Assisted Driving 383
15.2.3 Benefits for Active Safety Systems 384
15.2.4 Development Process for Active Safety 385
15.2.5 Future Requirements and Perspectives 386
15.3 Prospective Evaluation of the Effectiveness of Active Safety Systems and HAD Systems 388
15.3.1 Challenges 388
15.3.2 Variability at the Model Design 388
15.3.3 Evaluation of the Effectiveness 389
15.4 Conclusion 390
References 390
16 Functional Safety of Automated Driving Systems: Does ISO 26262 Meet the Challenges? 392
16.1 Introduction 392
16.1.1 From Driver Assistance to Highly Automated Driving Systems 393
16.1.2 Functional Safety According to ISO 26262 396
16.2 General Challenges of ADS 398
16.2.1 Increasing Complexity of ADS 399
16.2.2 Strict Requirements Concerning Availability and Reliability of ADS 401
16.3 Challenges to ADS Concerning Functional Safety 401
16.3.1 Vehicle Platform for Basic Driving Functions 402
16.3.2 From ADAS to ADS Functions 403
16.3.3 Share of Sensors and Actuators 403
16.3.4 From Many ECUs to Host ECUs 403
16.4 Importance of the Concept Phase 404
16.4.1 Item Definition 404
16.4.2 Hazard Analysis and Risk Assessment 405
16.4.3 Determination of ASIL and Safety Goals 406
16.4.4 Functional Safety Concept 408
16.4.4.1 Examples of FSC for Different ADS Levels 408
16.4.4.2 Vital Role of the Driver in the FSC 410
16.5 Supporting Methods to Handle Complexity of ADS 410
16.5.1 Model-Based Systems Engineering 411
16.5.2 Formal Verification by Contract-Based Design 412
16.5.3 Simulation and Co-simulation 414
16.6 Further Safety-Related Topics 415
16.6.1 Influence of Security on Safety Functions 415
16.6.2 Liability of ADS 416
16.6.3 Validation of ADS Functions 417
16.7 Conclusion 418
16.8 Acknowledgements 419
References 419
Part VI Validation and Testing of Automated Driving Functions 422
17 The New Role of Road Testing for the Safety Validation of Automated Vehicles 423
17.1 Introduction 423
17.2 The Goal of the Validation in Terms of Safety 424
17.3 Challenges for Safety Validation of Automated Vehicles Based on Road Testing 425
17.4 Challenges for New Approaches on Safety Validation 427
17.4.1 New Approaches on Safety Validation 427
17.4.2 Validation of Alternative Approaches by Road Testing 428
17.5 A Confession About the First Introduction of Automated Vehicles 429
17.6 Argumentation for Introduction of Automated Systems Motivated by Statistics 430
17.6.1 Universal Theory on a “Brave Introduction” of Automated Systems 432
17.6.1.1 User's Perspective 433
17.6.1.2 Society's Perspective 434
17.6.2 EXAMPLE: Introducing Highly Automated Driving on German Autobahn 436
17.7 Conclusion 437
References 438
18 Validation of Highly Automated Safe and Secure Systems 440
18.1 Introduction 440
18.2 Complexity of Automated Vehicles 442
18.3 Validation Challenges 443
18.4 Validation Concepts 447
18.5 Virtual Validation Environment 449
18.6 Conclusion 452
References 453
19 Testing and Validating Tactical Lane Change Behavior Planning for Automated Driving 454
19.1 Introduction 454
19.1.1 Motivation 454
19.1.2 Article Outline 456
19.2 Definition of Terms Scene, Situation, and Scenario 456
19.3 Background 458
19.4 Integrating Unit Tests, Situation-Based Open-Loop Testing, and Scenario-Based Closed-Loop Testing into the V-Model 460
19.4.1 Unit Tests 462
19.4.2 Situation-Based Open-Loop Testing 462
19.4.3 Scenario-Based Closed-Loop Testing 464
19.4.4 Real World Driving Tests 465
19.5 Case Study: Testing and Validating Tactical Lane Change Behavior Planning 466
19.5.1 Item Under Test: Behavior Planning for Lane Changes 466
19.5.2 Situation-Based Open-Loop Testing 467
19.5.3 Scenario-Based Closed-Loop Testing 467
19.6 Conclusions 473
References 473
20 Safety Performance Assessment of Assisted and Automated Driving in Traffic: Simulation as Knowledge Synthesis 475
20.1 Introduction 475
20.1.1 Driver Assistance and Automation 475
20.1.2 Assessment and Optimization of ADAS and ADF as Key Processes During the Development 477
20.2 Overall Safety Assessment 478
20.2.1 Safety and Economy 478
20.2.2 Conflicting Objectives in Vehicle and Traffic Safety 478
20.3 Design and Optimization of ADAS Using Virtual Experiments 481
20.3.1 Paradigm of Design of Virtual Experiments 481
20.3.2 Representation of Safety-Relevant Processes in Simulation 481
20.3.3 Knowledge Synthesis and Integration of Other Test Domains 482
20.3.4 Process Description for Assessing Pedestrian Protection 484
20.3.5 Simulation of ADAS Effectiveness 485
20.3.6 Interpretation of ADAS Effectiveness 487
20.4 New Challenges for Virtual Assessment of Automated Driving Functions 488
20.4.1 Impact of Automated Driving Functions on Safety-Related Processes in Traffic 488
20.4.2 Contributions to Safety Impact in Existing Risk Scenarios 488
20.4.3 Expanding the Spectrum of Safety-Relevant Scenarios 489
20.4.4 Philosophy and Procedural Approaches for Validation and Assessment of Automation 491
20.5 Conclusion and Outlook 493
References 495
21 From Controllability to Safety in Use: Safety Assessment of Driver Assistance Systems 497
21.1 Introduction 497
21.2 Controllability of Driver Assistance Systems 498
21.3 Safety in Use: A Holistic Consideration of the Driver, Vehicle, and Environment 500
21.3.1 Systematic Analysis of Safety in Use 501
21.3.2 Reference Values for Safety in Use 504
21.4 Informative Sources for the Creation of an Analysis of Safety in Use 507
21.4.1 Data Collection: Literature Review 509
21.4.2 Data Collection: Questionnaires 509
21.4.3 Data Collection: Studies in the Driving Simulator or in Real Vehicles 511
21.4.4 Data Collection: Observation of Traffic 513
21.4.5 Field Operational Test 515
21.5 Conclusion 518
References 519
22 Testing Autonomous and Highly Configurable Systems: Challenges and Feasible Solutions 521
22.1 Introduction 521
22.2 Related Research 523
22.3 Problem Definition 525
22.4 Combinatorial Testing for Autonomous Adaptive Systems 527
22.4.1 Combinatorial Testing 527
22.4.2 Test Oracles 530
22.4.3 The Automated Testing Methodology 531
22.5 Conclusion 533
References 534
Part VII A Sampling of Automated Driving Research Projects and Initiatives 535
European and National Projects 533
23 AdaptIVe: Automated Driving Applications and Technologies for Intelligent Vehicles 536
23.1 Project Overview 536
23.2 Technical Areas of AdaptIVe 537
23.2.1 Legal Aspects 537
23.2.2 Human–Vehicle Integration 538
23.2.3 Close-Distance Scenarios 539
23.2.4 Urban Scenarios 539
23.2.5 Highway Scenarios 540
23.2.6 Evaluation 541
23.3 Looking Ahead 541
24 When Autonomous Vehicles Are Introduced on a Larger Scale in the Road Transport System: The Drive Me Project 542
24.1 Introduction 542
24.2 Problem Definition 543
24.3 Measuring Probes: The Autonomous Vehicles 544
24.4 Safety 544
24.5 Traffic Flow 545
24.6 Energy Efficiency 546
24.7 Conclusion 546
References 546
25 Functional Safety and Evolvable Architectures for Autonomy 548
25.1 Introduction 548
25.2 Why It Is Harder to Show Functional Safety for Autonomous Vehicles 550
25.2.1 Item Definition 550
25.2.2 The Role of the Driver 551
25.3 How to Perform Hazard Analysis and Risk Assessment 552
25.3.1 Preliminary Feature Description 552
25.3.2 Situation Analysis and Hazard Identification 555
25.3.3 Find Dimensioning Hazardous Events 556
25.3.4 Function Refinement 556
25.3.5 Item Definition 556
25.3.6 Safety Goals, FSC, and TSC 557
25.3.7 Consequences for Safety Case and Assessment 557
25.4 How to Refine Safety Requirements 558
25.5 What Functional Architectures Fit Autonomous Driving 559
25.6 Conclusion 560
References 561
26 Challenges for Automated Cooperative Driving: The AutoNet2030 Approach 562
26.1 Introduction 562
26.2 Use Cases 563
26.3 Human Machine Interface 565
26.4 Cooperative Control 566
26.4.1 Distributed Graph-Based Convoy Control 566
26.4.2 Cooperative Intersection Management 567
26.5 Cooperative Sensing and Perception Layer 567
26.5.1 Configurable Perception Layer 567
26.5.2 V2X Communications for Automated Driving 568
26.5.3 Road Data Fusion Module 570
26.6 Conclusion and Outlook 570
References 570
27 Architecture and Safety for Autonomous HeavyVehicles: ARCHER 572
27.1 Summary of the Project 572
27.2 Background 573
27.3 State of the Art 574
27.4 Project Content 578
27.5 Project Targets 580
References 580
28 Affordable Safe and Secure Mobility Evolution 583
28.1 Mobility Systems Evolution 583
28.2 Objectives 585
28.3 Expected Outcomes 586
Reference 588
29 UFO: Ultraflat Overrunable Robot for Experimental ADAS Testing 589
29.1 Introduction 589
29.2 Structure of the UFO Platform 589
29.3 Communication Infrastructure 591
29.4 Definition of Test Scenarios 592
29.5 Summary 593
References 594
30 Intelligent Transport Systems: The Trials Making Smart Mobility a Reality 595
30.1 ITS Corridor: Austria, Germany and The Netherlands 596
30.2 Helmond 597
30.3 Hamburg 598
European and National Initiatives 598
31 A Sampling of Automated Driving Research Projects and Initiatives (EC Funded, National) 601
31.1 Introduction 601
31.2 Mission of ARTEMIS-IA 602
31.3 Structure of ARTEMIS-IA 602
31.4 Automated Driving and ARTEMIS-IA 603
References 606
32 ERTRAC: The European Road Transport Research Advisory Council 607
32.1 Introduction 607
32.2 Mission of ERTRAC 608
32.3 Structure of ERTRAC 609
32.4 Automated Driving Roadmap of ERTRAC 610
Reference 610
33 SafeTRANS: Safety in Transportation Systems 611
33.1 Introduction 611
33.2 R& D Strategies and Roadmaps
33.3 Working Group on Highly Autonomous Systems: Safety, Testing, and Development Process 613
33.4 Sustainability and Standardization 614
33.5 Conclusion 615
References 615
34 A3PS: Austrian Association for Advanced Propulsion Systems 616
34.1 Objectives and Tasks of A3PS 616
34.2 ADAS in the Technology Roadmap of A3PS 617