Zhou / Shen / Yong Theories and Practices of Self-Driving Vehicles

1. Auflage 2022
ISBN: 978-0-323-99449-1
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, 342 Seiten

Self-driving vehicles are a rapidly growing area of research and expertise. Theories and Practice of Self-Driving Vehicles presents a comprehensive introduction to the technology of self driving vehicles across the three domains of perception, planning and control. The title systematically introduces vehicle systems from principles to practice, including basic knowledge of ROS programming, machine and deep learning, as well as basic modules such as environmental perception and sensor fusion. The book introduces advanced control algorithms as well as important areas of new research. This title offers engineers, technicians and students an accessible handbook to the entire stack of technology in a self-driving vehicle. Theories and Practice of Self-Driving Vehicles presents an introduction to self-driving vehicle technology from principles to practice. Ten chapters cover the full stack of driverless technology for a self-driving vehicle. Written by two authors experienced in both industry and research, this book offers an accessible and systematic introduction to self-driving vehicle technology. - Provides a comprehensive introduction to the technology stack of a self-driving vehicle - Covers the three domains of perception, planning and control - Offers foundational theory and best practices - Introduces advanced control algorithms and high-potential areas of new research - Gives engineers, technicians and students an accessible handbook to self-driving vehicle technology and applications

Qingguo Zhou is Professor at Lanzhou University and Deputy Director of the Engineering Research Center for Open Source Software and Real-Time Systems, at the Ministry of Education, China. He is also the Director of the School of Computer Science and Engineering and the Embedded System Laboratory at Lanzhou University. His research focuses on intelligent driving, AI, embedded and real-time systems. He has published widely.

Autoren/Hrsg.

Zhou, Qingguo

Shen, Zebang

Yong, Binbin

Zhao, Rui

Zhi, Peng

Inhaltsverzeichnis

1;Front Cover;1
2;THEORIES AND PRACTICES OF SELF-DRIVING VEHICLES;2
3;THEORIES AND PRACTICES OF SELF-DRIVING VEHICLES;4
4;Copyright;5
5;Contents;6
6;Contributors;10
7;1 - First acquaintance with unmanned vehicles;12
7.1;1.1 What are unmanned vehicles?;13
7.1.1;1.1.1 Classification standards for unmanned vehicles;13
7.1.2;1.1.2 How difficult is the implementation of unmanned vehicles?;16
7.2;1.2 Why do we need unmanned vehicles?;18
7.2.1;1.2.1 Improvement of road traffic safety;18
7.2.2;1.2.2 Alleviation of urban traffic congestion;19
7.2.3;1.2.3 Improvement of travel efficiency;19
7.2.4;1.2.4 Lowering the threshold for drivers;20
7.3;1.3 Basic framework of the unmanned vehicle system;20
7.3.1;1.3.1 Environmental perception;21
7.3.2;1.3.2 Localization;24
7.3.3;1.3.3 Mission planning;25
7.3.4;1.3.4 Behavior planning;27
7.3.5;1.3.5 Motion planning;28
7.3.6;1.3.6 Control system;29
7.3.7;1.3.7 Summary;32
7.4;1.4 Development environment configuration;33
7.4.1;1.4.1 Simple environment installation;33
7.4.2;1.4.2 Install the robot operating system (ROS);34
7.4.3;1.4.3 Install OpenCV;35
7.5;References;36
8;2 - Introduction to robot operating system;38
8.1;2.1 ROS introduction;39
8.1.1;2.1.1 Brief introduction to ROS;39
8.1.1.1;2.1.1.1 What is ROS?;39
8.1.1.2;2.1.1.2 History of ROS;39
8.1.1.3;2.1.1.3 Features of ROS;40
8.1.2;2.1.2 Concept of ROS;40
8.1.2.1;2.1.2.1 Master;40
8.1.2.2;2.1.2.2 Node;41
8.1.2.3;2.1.2.3 Topic;41
8.1.2.4;2.1.2.4 Message;43
8.1.3;2.1.3 Catkin create system;43
8.1.4;2.1.4 Project organization structure in ROS;44
8.1.4.1;2.1.4.1 CMakeLists.txt;45
8.1.5;2.1.5 Practice based on husky simulator;46
8.1.6;2.1.6 Basic ROS programming;50
8.1.6.1;2.1.6.1 ROS C++ client library (roscpp);50
8.1.6.1.1;2.1.6.1.1 Node handle;52
8.1.6.1.2;2.1.6.1.2 ROS logging method;53
8.1.6.2;2.1.6.2 Write simple publish and subscribe code;53
8.1.6.2.1;2.1.6.2.1 Object-oriented node coding;55
8.1.6.3;2.1.6.3 Parameter services in ROS;55
8.1.6.4;2.1.6.4 Small case based on husky robot;56
8.1.7;2.1.7 ROS services;64
8.1.8;2.1.8 ROS action;68
8.1.9;2.1.9 Common tools in ROS;68
8.1.9.1;2.1.9.1 Rviz;69
8.1.9.2;2.1.9.2 rqt;69
8.1.9.3;2.1.9.3 TF coordinate conversion system;70
8.1.9.4;2.1.9.4 URDF and SDF;72
8.2;References;73
9;3 - Localization for unmanned vehicle;74
9.1;3.1 Principle of achieving localization;74
9.2;3.2 ICP algorithm;76
9.3;3.3 Normal distribution transform;83
9.3.1;3.3.1 Introduction to NDT algorithm;83
9.3.2;3.3.2 Basic steps of NDT algorithm;84
9.3.3;3.3.3 Advantages of NDT algorithm;85
9.3.4;3.3.4 Algorithm example;86
9.4;3.4 Localization system based on global positioning system (GPS) + inertial navigation system (INS);92
9.4.1;3.4.1 Localization principle;93
9.4.2;3.4.2 Localization fusion of different sensors;95
9.5;3.5 SLAM-based localization system;98
9.5.1;3.5.1 SLAM localization principle;99
9.5.2;3.5.2 SLAM applications;101
9.6;References;104
10;4 - State estimation and sensor fusion;106
10.1;4.1 Kalman filter and state estimation;107
10.1.1;4.1.1 What is the Kalman filter?;107
10.1.2;4.1.2 Kalman filter;107
10.1.2.1;4.1.2.1 Status forecast;108
10.1.2.2;4.1.2.2 Calculation of forecast error;109
10.1.2.3;4.1.2.3 Measurement error;109
10.1.2.4;4.1.2.4 Calculation of Kalman gain;110
10.1.2.5;4.1.2.5 Calculation of optimal estimate;110
10.1.2.6;4.1.2.6 Calculation of the error of the optimal estimate;111
10.1.3;4.1.3 Kalman filter in autonomous vehicle sensing module;113
10.1.3.1;4.1.3.1 Sensors for autonomous vehicle perception module;113
10.1.3.2;4.1.3.2 Kalman filter-based pedestrian localization estimation;113
10.1.3.3;4.1.3.3 Kalman filter pedestrian state estimation in Python;116
10.2;4.2 Advanced motion modeling and EKF;126
10.2.1;4.2.1 Advanced motion models for vehicle tracking;126
10.2.2;4.2.2 EKF;129
10.2.2.1;4.2.2.1 Jacobi matrix;129
10.2.2.2;4.2.2.2 Process noise;132
10.2.2.3;4.2.2.3 Measurement;135
10.2.2.4;4.2.2.4 Python implementation;137
10.3;4.3 UKF;149
10.3.1;4.3.1 Movement model;150
10.3.2;4.3.2 Nonlinear processing/measurement models;151
10.3.3;4.3.3 Lossless transformation;151
10.3.4;4.3.4 Projections;152
10.3.4.1;4.3.4.1 Prediction of sigma point;153
10.3.4.2;4.3.4.2 Predicted mean and variance;153
10.3.5;4.3.5 Measurement updates;154
10.3.5.1;4.3.5.1 Update status;155
10.3.6;4.3.6 Summary;155
10.4;References;155
11;5 - Introduction of machine learning and neural networks;158
11.1;5.1 Basic concepts of machine learning;159
11.2;5.2 Supervised learning;162
11.2.1;5.2.1 Empirical risk minimization;162
11.2.2;5.2.2 Overfitting and underfitting;163
11.2.3;5.2.3 ``Certain algorithm''-gradient descent algorithm;166
11.2.4;5.2.4 Summary;167
11.3;5.3 Fundamentals of neural network;168
11.3.1;5.3.1 Basic structure of the neural network;169
11.3.2;5.3.2 Unlimited capacity-fitting arbitrary functions;171
11.3.3;5.3.3 Forward transmission;173
11.3.4;5.3.4 Stochastic gradient descent;175
11.4;5.4 Using Keras to implement the neural network;176
11.4.1;5.4.1 Data preparation;176
11.4.2;5.4.2 A small change in three-layer neural network-deep feedforward neural network;182
11.4.3;5.4.3 Summary;186
11.5;References;186
12;6 - Deep learning and visual perception;188
12.1;6.1 Deep feedforward neural networks-why is it necessary to be deep?;189
12.1.1;6.1.1 Efficiency of model training under big data;189
12.1.2;6.1.2 Representation learning;190
12.2;6.2 Regularization technology applied to deep neural networks;191
12.2.1;6.2.1 Data augmentation;191
12.2.2;6.2.2 Early stopping;194
12.2.3;6.2.3 Parameter normalization penalties;195
12.2.4;6.2.4 Dropout;196
12.3;6.3 Actual combat-traffic sign recognition;198
12.3.1;6.3.1 Belgium traffic sign dataset;198
12.3.2;6.3.2 Data preprocessing;204
12.3.3;6.3.3 Leverage Keras to construct and train a deep feedforward network;206
12.4;6.4 Introduction to convolutional neural networks;207
12.4.1;6.4.1 What is convolution and the motivation for convolution;208
12.4.2;6.4.2 Sparse interactions;210
12.4.3;6.4.3 Parameter sharing;213
12.4.4;6.4.4 Equivariant representations;213
12.4.5;6.4.5 Convolutional neural network;213
12.4.6;6.4.6 Some details of convolution;215
12.5;6.5 Vehicle detection based on YOLO2;217
12.5.1;6.5.1 Pretrained classification network;218
12.5.2;6.5.2 Train the detection network;219
12.5.3;6.5.3 Loss function of YOLO;220
12.5.4;6.5.4 Test;221
12.5.5;6.5.5 Vehicle and pedestrian detection based on YOLO;221
12.6;References;227
13;7 - Transfer learning and end-to-end self-driving;228
13.1;7.1 Transfer learning;229
13.2;7.2 End-to-end selfdriving;231
13.3;7.3 End-to-end selfdriving simulation;232
13.3.1;7.3.1 Selection of the simulator;232
13.3.2;7.3.2 Data collection and processing;233
13.3.3;7.3.3 Construction of the deep neural network model;234
13.3.3.1;7.3.3.1 LeNet deep selfdriving model;235
13.3.3.2;7.3.3.2 NVIDIA deep selfdriving model;237
13.4;7.4 Summary of this chapter;239
13.5;References;240
14;8 - Getting started with self-driving planning;242
14.1;8.1 A? algorithm;243
14.1.1;8.1.1 Directed graph;243
14.1.2;8.1.2 Breadth-first search (BFS) algorithm;244
14.1.3;8.1.3 Data structure;246
14.1.4;8.1.4 How to generate a route;247
14.1.5;8.1.5 Directional search (heuristic);249
14.1.6;8.1.6 Dijkstra algorithm;250
14.1.7;8.1.7 A? algorithm;251
14.2;8.2 Hierarchical finite state machine (HFSM) and autonomous vehicle behavior planning;252
14.2.1;8.2.1 Design criteria for decision-making plan system of autonomous vehicles;253
14.2.2;8.2.2 FSM;253
14.2.3;8.2.3 Hierarchical FSM;255
14.2.4;8.2.4 Use of state machines in behavior planning;256
14.3;8.3 Autonomous vehicle route generation based on free boundary cubic spline interpolation;258
14.3.1;8.3.1 Cubic spline interpolation;258
14.3.2;8.3.2 Cubic spline interpolation algorithm;262
14.3.3;8.3.3 Using Python to implement cubic spline interpolation for path generation;264
14.4;8.4 Motion planning method of the autonomous vehicle based on Frenet optimization trajectory;266
14.4.1;8.4.1 Why use the Frenet coordinate system;268
14.4.2;8.4.2 Jerk minimization and polynomial solution of fifth degree trajectory;270
14.4.3;8.4.3 Collision avoidance;275
14.4.4;8.4.4 Example of motion planning for autonomous vehicles based on Frenet optimization trajectory;276
14.5;References;283
15;9 - Vehicle model and advanced control;284
15.1;9.1 Kinematic bicycle model and dynamic bicycle model;284
15.1.1;9.1.1 Bicycle model;285
15.1.2;9.1.2 Kinematic bicycle model;287
15.1.3;9.1.3 Dynamic bicycle model;288
15.2;9.2 Rudiments of autonomous vehicle control;289
15.2.1;9.2.1 Need for control theory;289
15.2.2;9.2.2 PID control algorithm;290
15.2.2.1;9.2.2.1 Proportional control;291
15.2.2.2;9.2.2.2 Proportional and derivative control;293
15.3;9.3 MPC based on kinematic model;302
15.3.1;9.3.1 Applying PID controller forwards to steering control;302
15.3.2;9.3.2 Predictive model;302
15.3.3;9.3.3 Online optimal loop based on time series;304
15.3.4;9.3.4 Feedback correction;305
15.4;9.4 Trajectory tracking;306
15.5;References;316
16;10 - Deep reinforcement learning and application in self-driving;318
16.1;10.1 Overview of reinforcement learning;319
16.2;10.2 Reinforcement learning;320
16.2.1;10.2.1 Markov decision process;320
16.2.2;10.2.2 Constituent elements;321
16.2.2.1;10.2.2.1 Policy;321
16.2.2.2;10.2.2.2 Reward;322
16.2.3;10.2.3 Value function;322
16.3;10.3 Approximate value function;325
16.4;10.4 Deep Q network algorithm;326
16.4.1;10.4.1 Q learning algorithm;326
16.4.2;10.4.2 DQN algorithm;326
16.4.2.1;10.4.2.1 Reward function;327
16.4.2.2;10.4.2.2 Objective function;328
16.5;10.5 Policy gradient;330
16.6;10.6 Deep deterministic policy gradient and TORCS game control;330
16.6.1;10.6.1 About TORCS game;330
16.6.2;10.6.2 TORCS game environment installation;331
16.6.3;10.6.3 Deep deterministic strategy gradient algorithm;333
16.6.3.1;10.6.3.1 Theory;333
16.6.3.2;10.6.3.2 Reward function setting;334
16.6.3.3;10.6.3.3 Running program;335
16.7;10.7 Summary;336
16.8;References;336
17;Index;338
17.1;A;338
17.2;B;338
17.3;C;338
17.4;D;339
17.5;E;340
17.6;F;340
17.7;G;340
17.8;H;340
17.9;I;341
17.10;J;341
17.11;K;341
17.12;L;341
17.13;M;341
17.14;N;342
17.15;O;342
17.16;P;342
17.17;R;343
17.18;S;344
17.19;T;344
17.20;U;344
17.21;V;345
17.22;W;345
17.23;Y;345
18;Back Cover;346