The day will soon come when you will be able to verbally communicate with a vehicle and instruct it to drive to a location. The car will navigate through street traffic and take you to your destination without additional instruction or effort on your part. Today, this scenario is still in the future, but the automotive industry is racing toward the finish line to have automated driving vehicles deployed on our roads.
ADAS and Automated Driving: A Practical Approach to Verification and Validation focuses on how automated driving systems (ADS) can be developed from concept to a product on the market for widescale public use. It covers practically viable approaches, methods, and techniques with examples from multiple production programs across different organizations. The author provides an overview of the various Advanced Driver Assistance Systems (ADAS) and ADS currently being developed and installed in vehicles.
The technology needed for large-scale production and public use of fully autonomous vehicles is still under development, and the creation of such technology is a highly innovative area of the automotive industry. This text is a comprehensive reference for anyone interested in a career focused on the verification and validation of ADAS and ADS. The examples included in the volume provide the reader foundational knowledge and follow best and proven practices from the industry. Using the information in ADAS and Automated Driving, you can kick start your career in the field of ADAS and ADS.
Author(s): Plato Pathrose
Publisher: SAE International
Year: 2022
Language: English
Pages: 277
City: Warrendale
Front Cover
Title Page
Copyright Page
Dedication
Contents
Foreword
Introduction
About This Book
Assumptions
Acknowledgments
Chapter 1 Introduction to Advanced Driver Assistance Systems and Automated Driving
1.1 Sense Organs of a Vehicle
1.1.1 Camera
1.1.2 Radar
1.1.3 Lidar
1.1.4 Ultrasonic Sensors
1.1.5 Inertial Measurement Unit Sensors
1.1.6 High-Definition Maps
1.2 ADAS and Automated Driving
1.2.1 Highway Assist and Traffic Jam Assist (Level 2)
1.2.2 Remote Parking (Level 2)
1.2.3 Traffic Jam Chauffeur (Level 3)
1.2.4 Highway Chauffeur (Level 3)
1.2.5 Urban and Suburban Pilot (Level 4)
1.2.6 Highway Autopilot (Level 4)
1.2.7 Valet Parking (Level 4)
1.3 Level 5: Full Automation
1.4 Operational Design Domain
1.5 Dynamic Driving Task
1.6 Object and Event Detection and Response
1.7 Summary
References
Chapter 2 Design Approaches for Automated Driving Systems
2.1 Product Development
2.2 Distributed Architecture versus Centralized Architecture
2.3 Developing an Automated Driving System
2.4 Requirement Elicitation
2.5 Quality Function Deployment
2.6 Designing a Robust Product
2.7 Failure Mode and Effects Analysis
2.8 Summary
References
Chapter 3 Different Test Approaches
3.1 Verification and Validation
3.2 Agility in Verification and Validation
3.3 Different Levels of Testing— A Reference from V-Model
3.4 Defects at Different Levels of Testing
3.5 Simulation and Testing
3.5.1 Model-in-the-Loop Simulation
3.5.2 Software-in-the-Loop Simulation
3.5.3 Hardware-in-the-Loop Simulation
3.5.4 Driver-in-the-Loop Simulation
3.5.5 Vehicle-in-the-Loop Simulation
3.6 Summary
References
Chapter 4 Scenario-Based Testing
4.1 Scenario Elicitation, Description, and Structuring
4.2 Scenario Implementation and Parameterization
4.3 Scenario-Based Simulation and Testing
4.4 Scenario-Based Testing at Different Levels
4.5 Scenario Database Management
4.6 Automation in Scenario-Based Testing
4.7 Summary
References
Chapter 5 Simulation Environment for ADAS and Automated Driving Systems
5.1 Simulation Tool Selection
5.2 Co-simulation in Testing
5.3 General Qualification of Simulation Environment
5.4 Limitations of Simulation Tools Used in ADAS and Automated Driving
5.5 Summary
References
Chapter 6 Ground Truth Generation and Testing Neural Network-Based Detection
6.1 Introduction to Data-Driven Software Development
6.2 Data Annotation and Dataset Generation
6.3 Metric for Detection Quality Evaluation
6.4 Evaluating KPIs for Detection Algorithm
6.4.1 Preconditions for Sample Data Collection
6.4.2 Data and Data Types
6.4.3 Performance Evaluation (KPI Measurement)
6.4.3.1 Detection Evaluation on a Single Frame (Detection Performance)
6.4.3.2 Detection Evaluation on Complete Ground Truth Dataset (Detection Quality)
6.4.3.3 Detection Evaluation Using Noise Variants as Input (Detection Performance and Quality)
6.4.3.4 Detection Evaluation in the Vehicle (Detection Performance)
6.5 Different Acceptance Quality for Detection Algorithms
6.6 Challenges in Measuring Quality of Object Detection
6.7 Summary
References
Chapter 7 Testing and Qualification of Perception Software
7.1 Overview of Automated Driving Systems
7.2 Perception—An Architecture Overview
7.3 Different Methods for Perception Software Testing
7.4 Methods for Evaluating Perception Software Components
7.4.1 Evaluation of Static and Dynamic Object Fusion
7.4.2 Evaluation of Grid Fusion
7.4.3 Evaluation of Localization
7.4.4 Evaluation of Prediction Algorithms
7.5 Measuring Performance and Quality of Perception Software
7.5.1 Preconditions for Measurements
7.5.2 Data and Data Types
7.5.3 Performance Evaluation (KPI Measurement)
7.6 Testing Robustness of the Perception Software
7.7 Challenges in the Measurement and Evaluation of Perception
7.8 Summary
References
Chapter 8 Calibration of ADAS and Automated Driving Features
8.1 Calibration—An Overview Based on Ideality Equation
8.2 Common Types of Calibration in an Automated Driving System
8.2.1 End of Line (EoL) Calibration
8.2.2 Service Calibration
8.2.3 Online Calibration
8.2.4 Functional Calibration
8.3 Calibration of ADAS and Automated Driving Features
8.4 Calibration Environment for Automated Driving Vehicles
8.5 Calibration over Diagnostics Interface
8.6 Summary
References
Chapter 9 Introduction to Functional Safety and Cybersecurity Testing
9.1 Functional Safety and Cybersecurity in Automotive
9.2 Safety Qualification of Tools and Toolchain
9.3 An Overview of Functional Safety Testing
9.4 Fault Injection Testing Using Diagnostics
9.5 Safety Testing of Artificial Neural Networks—An Overview
9.6 An Overview of Cybersecurity Testing
9.7 Summary
References
Chapter 10 Verification and Validation Strategy
10.1 Test-Driven Development and Feature-Driven Development
10.2 Purpose of Test Design and Test Depth
10.3 Developing a Test Suite
10.4 Test Process
10.5 Testing in the Vehicle
10.6 Summary
References
Chapter 11 Acceptance Criteria and Maturity Evaluation
11.1 Need for Acceptance Criteria
11.2 Defining Maturity of the System and Features
11.3 Maturity Evaluation for the System
11.4 Maturity Evaluation for the Features
11.5 Vehicle Testing and Feature Maturity Evaluation
11.6 Case Study on How Various ADAS Features Are Deployed
11.7 Summary
References
Chapter 12 Data Flow and Management in Automated Driving
12.1 Importance of Data in Automated Driving
12.2 Types of Data Collected
12.3 Data Acquisition Strategy and Data Loggers
12.4 Data Reuse Strategy
12.5 Data Analysis and Data Flow
12.6 Data Storage and Management— A Case Study
12.7 Challenges in Data Acquisition and Management
12.8 Summary
References
Chapter 13 Challenges and Gaps in Testing Automated Driving Features
13.1 Challenges due to Infrastructure Quality
13.2 Challenges in the Design of Automated Driving Systems
13.3 Challenges in Performing Simulation-Based Testing
13.4 Challenges in Laboratory-Based Tests and Vehicle Tests
13.5 Challenges in Using AI
13.6 Challenges in Scenario-Based Testing
13.7 Challenges in Testing for Functional Safety and Cybersecurity
13.8 Challenges with Legal Aspects, Liabilities and Its Economic Impacts
13.9 Summary
References
Index
About the Author
Back Cover
