Robust Environmental Perception and Reliability Control for Intelligent Vehicles (Recent Advancements in Connected Autonomous Vehicle Technologies, 4)

✍ Scribed by Huihui Pan, Jue Wang, Xinghu Yu, Weichao Sun, Huijun Gao

• Publisher: Springer
• Year: 2023
• Tongue: English
• Leaves: 308
• Edition: 1
• Category: Library

✦ Synopsis

This book presents the most recent state-of-the-art algorithms on robust environmental perception and reliability control for intelligent vehicle systems. By integrating object detection, semantic segmentation, trajectory prediction, multi-object tracking, multi-sensor fusion, and reliability control in a systematic way, this book is aimed at guaranteeing that intelligent vehicles can run safely in complex road traffic scenes.

• Adopts the multi-sensor data fusion-based neural networks to environmental perception fault tolerance algorithms, solving the problem of perception reliability when some sensors fail by using data redundancy.
• Presents the camera-based monocular approach to implement the robust perception tasks, which introduces sequential feature association and depth hint augmentation, and introduces seven adaptive methods.
• Proposes efficient and robust semantic segmentation of traffic scenes through real-time deep dual-resolution networks and representation separation of vision transformers.
• Focuses on trajectory prediction and proposes phased and progressive trajectory prediction methods that is more consistent with human psychological characteristics, which is able to take both social interactions and personal intentions into account.

Puts forward methods based on conditional random field and multi-task segmentation learning to solve the robust multi-object tracking problem for environment perception in autonomous vehicle scenarios.

• Presents the novel reliability control strategies of intelligent vehicles to optimize the dynamic tracking performance and investigates the completely unknown autonomous vehicle tracking issues with actuator faults.

✦ Table of Contents

Preface
Acknowledgements
Contents
1 Background
1.1 Introduction
1.2 Robust Environmental Perception
1.2.1 Multi-sensor Data Fusion
1.2.2 Monocular 3D Object Detection
1.2.3 Semantic Segmentation
1.2.4 Trajectory Prediction
1.2.5 Multi-object Tracking
1.3 Reliability Control
1.4 Preview of Chapters
References
2 Robust Environmental Perception of Multi-sensor Data Fusion
2.1 Keypoint Estimation with Robust Model
2.1.1 Problem Formulation
2.1.2 Structure of the Proposed RGKCNet
2.1.3 Experimental Results
2.1.4 Conclusion
2.2 Multiple Cameras Fusion Based Object Detection for Sensor …
2.2.1 Data Fusion Framework
2.2.2 Design of Region Proposal Networks
2.2.3 Fault Diagnosis and Avoidance Mechanism
2.2.4 Network Performance Evaluation
2.2.5 Conclusion
2.3 A Multi-phase Camera-LiDAR Fusion Network for Robust 3D …
2.3.1 The Multi-phase Fusion Scheme
2.3.2 Experimental Verifications
2.3.3 Ablation Study
2.3.4 Conclusion
References
3 Robust Environmental Perception of Monocular 3D Object Detection
3.1 Robust Detection with Sequential Feature and Depth Hint
3.1.1 Design of Robust Monocular 3D Detection Network
3.1.2 Experimental Results
3.1.3 Conclusion
3.2 Adaptive Methods in Robust Monocular 3D Detection
3.2.1 Design of Adaptive Monocular 3D Detection Network
3.2.2 Experimental Verifications
3.2.3 Conclusion
References
4 Robust Environmental Perception of Semantic Segmentation
4.1 Deep Dual-Resolution Networks for Fast and Robust Semantic Segmentation
4.1.1 Introduction
4.1.2 Deep Dual-Resolution Networks
4.1.3 Experimental Results
4.1.4 Conclusion
4.2 Robust Semantic Segmentation with Representation Separation
4.2.1 Introduction
4.2.2 Representation Separation of Vision Transformers
4.2.3 Experimental Results
4.2.4 Conclusion
References
5 Robust Environmental Perception of Trajectory Prediction
5.1 Transformers for Trajectory Prediction
5.1.1 The Proposed Model
5.1.2 Results
5.1.3 Conclusion
5.2 Multi-level CNNs
5.2.1 Overview for Multi-level CNNs
5.2.2 Verifications
5.2.3 Conclusion
References
6 Robust Environmental Perception of Multi-object Tracking
6.1 Robust Perceptual Tracking Oriented to Tracklet Inactivation
6.1.1 Overview of Multi-object Tracking Research
6.1.2 Methods of Multi-object Tracking Perception
6.1.3 Tracking Framework Based on Conditional Random Fields
6.1.4 Implementation of Tracking Perception
6.1.5 Performance Analysis of Multi-target Tracking
6.1.6 Conclusion
6.2 Learning Multi-task for Tracking Based on Segmentation
6.2.1 Segmentation Based Multi-object Tracking
6.2.2 Tracking in Combination with Multi-task Learning
6.2.3 Robust Real-Time YOLACT Tracking Architecture
6.2.4 Metrics and Comparison of Tracking Datasets
6.2.5 Conclusion
References
7 Reliability Control of Intelligent Vehicles
7.1 Dual-Loop ADP Fault-Tolerant Trajectory Control for Intelligent Vehicles
7.1.1 System Model
7.1.2 Fault-Tolerant Tracking Controller
7.1.3 Simulation Verification
7.1.4 Conclusion
7.2 Fault-Tolerant Tracking Control for Unknown Intelligent Vehicles
7.2.1 Problem Formulation
7.2.2 Model-Free Fault-Tolerant Controller
7.2.3 Simulation Verification
7.2.4 Conclusion
7.3 Finite-Time Fault-Tolerant Integrated Motion Control for Intelligent Vehicles
7.3.1 Problem Statement
7.3.2 Fault-Tolerant Integrated Motion Control Scheme
7.3.3 Simulation Analysis
7.3.4 Conclusion
References