Unmanned Driving Systems for Smart Trains

Title-page_2020_Unmanned-Driving-Systems-for-Smart-Trains
Unmanned Driving Systems for Smart Trains
Copyright_2021_Unmanned-Driving-Systems-for-Smart-Trains
Copyright
Contents_2020_Unmanned-Driving-Systems-for-Smart-Trains
Contents
List-of-figures_2021_Unmanned-Driving-Systems-for-Smart-Trains
List of figures
List-of-tables_2021_Unmanned-Driving-Systems-for-Smart-Trains
List of tables
Preface_2021_Unmanned-Driving-Systems-for-Smart-Trains
Preface
Acknowledgments_2021_Unmanned-Driving-Systems-for-Smart-Trains
Acknowledgments
Abbreviation-List_2021_Unmanned-Driving-Systems-for-Smart-Trains
Abbreviation List
Chapter-1---Introduction-of-the-train-unma_2021_Unmanned-Driving-Systems-for
1 Introduction of the train unmanned driving system
1.1 Overview of the train unmanned driving system
1.1.1 History of unmanned driving technology
1.1.2 The operation levels of automatic trains
1.1.3 The main functions and development of unmanned driving trains
1.1.4 The application fields of artificial intelligence in unmanned driving technology
1.1.4.1 Application of artificial intelligence in the transportation industry
1.1.4.2 Analysis of key technologies of artificial intelligence in unmanned driving trains
1.1.4.2.1 Control technology of rail trains based on deep reinforcement learning
1.1.4.2.2 Rail vehicle configuration technology based on big data analysis
1.1.4.2.3 Vehicle internet of things based on 5G communication and cloud computing
1.1.5 The development of unmanned driving in China
1.1.6 Achievements and developing trends with the cooperative initiative of “The Belt and Road”
1.2 The key issues of the unmanned driving system
1.2.1 The main control systems of unmanned drive technology
1.2.2 The scenario description of unmanned driving
1.2.3 The information integration of train scheduling
1.2.4 Important equipment of unmanned driving
1.2.5 Energy-saving methods for higher performance and lower consumption
1.2.6 Detection technology
1.2.7 Systematic reliability
1.2.8 Design of safety assessment system
1.2.9 Intelligent maintenance and operation
1.3 The scope of the book
1.3.1 The subsystems and performance evaluation system of unmanned driving
1.3.2 The main training algorithms
1.3.3 Research of main control parameters
1.3.4 Data mining and processing
1.3.5 Research of energy saving
1.3.6 The establishment of the simulation platform of algorithms
References
Chapter-2---Train-unmanned-driving-system-and-its_2021_Unmanned-Driving-Syst
2 Train unmanned driving system and its comprehensive performance evaluation system
2.1 Overview of automatic train operation/automatic train protection/automatic train supervision systems
2.1.1 The development of the automatic train control system
2.1.1.1 The historical process of the automatic train control system
2.1.1.1.1 The advantages of the automatic train control system
2.1.1.1.2 The research and development of a typical automatic train control system
2.1.1.2 The historical process of the automatic train operation system
2.1.1.3 The historical process of the automatic train protection system
2.1.1.4 The historical process of the automatic train supervision system
2.1.2 The structure and function of automatic train control systems
2.1.2.1 The structure and function of automatic train operation
2.1.2.1.1 The structure of the automatic train operation system
2.1.2.1.2 The function of the automatic train operation system
2.1.2.2 The structure and function of automatic train protection
2.1.2.2.1 The structure of the automatic train protection system
2.1.2.2.2 The function of the automatic train protection system
2.1.2.3 The structure and function of automatic train supervision
2.1.2.3.1 The structure of the automatic train supervision system
2.1.2.3.2 The function of the automatic train supervision system
2.1.3 The application of automatic train control systems
2.1.3.1 The application of the communications-based train control system in urban rail transit
2.1.3.2 Typical communications-based train control systems
2.1.3.2.1 Seltrac communications-based train control system
2.1.3.2.2 URBALIS communications-based train control systems
2.1.3.3 The application of the Chinese train control system 2+automatic train operation system
2.2 The performance indices of the train unmanned driving system
2.2.1 The performance indices of the automatic train operation system
2.2.1.1 Security
2.2.1.2 Traceability
2.2.1.3 Punctuality
2.2.1.4 Parking accuracy
2.2.1.5 Ride comfort
2.2.1.6 Energy saving
2.2.1.7 Traction brake switching frequency
2.2.1.8 Steady running speed
2.2.2 The performance indices of the automatic train protection system
2.2.2.1 The factors of the safety protection distance
2.2.2.1.1 Real-time speed and precalculating distance
2.2.2.1.2 Initial speed
2.2.2.1.3 Reaction time
2.2.2.2 The design principle of the safe protection distance
2.2.2.3 The calculation of the safety protection distance
2.2.3 The performance indices of the automatic train supervision system
2.3 The comprehensive performance evaluation methods of the train unmanned driving system
2.3.1 Comprehensive evaluation function
2.3.1.1 Principle of weight determination
2.3.1.2 Theoretical basis of the analytic hierarchy process
2.3.2 Analysis of automatic train operation hierarchical structure
2.3.2.1 Automatic train operation performance index confirmation study
2.3.2.2 Objective weight determination based on entropy
2.3.2.3 Subjective weight determination based on the analytic hierarchy process
2.3.3 Comprehensive weight determination method based on analytic hierarchy process-entropy
References
Chapter-3---Train-unmanned-driving-algorithm-bas_2021_Unmanned-Driving-Syste
3 Train unmanned driving algorithm based on reasoning and learning strategy
3.1 The current status and technical progress of train unmanned controlling algorithm
3.2 The connotation and composition of train unmanned driving algorithm
3.2.1 Research on the speed control of railway vehicles
3.2.1.1 Research on the modeling of driverless trains
3.2.1.2 Research on optimization of traction target curve for unmanned train
3.2.1.3 Research on speed tracking control of unmanned trains
3.2.2 Study on railway vehicle navigation system
3.2.2.1 Sensing
3.2.2.2 Perception
3.2.2.2.1 Positioning
3.2.2.2.2 Object recognition and tracking
3.2.2.3 Decision
3.2.3 Study on railway vehicle path planning
3.2.3.1 Global path planning algorithm
3.2.3.2 Local path planning algorithm
3.2.4 Study on target detection of railway vehicles
3.3 Calculation process and analysis of train unmanned driving algorithm
3.3.1 Positioning and navigation algorithm
3.3.1.1 Satellite positioning based on auxiliary augmentation
3.3.1.2 Location based on dead reckoning
3.3.1.3 Inertial navigation
3.3.1.4 Visual simultaneous localization and mapping
3.3.1.5 LiDAR simultaneous localization and mapping
3.3.1.6 Positioning technology based on beacon guidance
3.3.2 Path planning algorithm
3.3.2.1 Traditional algorithm
3.3.2.1.1 Graph search–based path planning algorithm
3.3.2.1.2 The sampling-based path planning algorithm
3.3.2.2 Intelligent optimization algorithm
3.3.2.2.1 Ant colony algorithm
3.3.2.2.2 Tentacle algorithm
3.3.2.2.3 Intelligent water drops algorithm
3.3.2.3 Algorithms based on reinforcement learning
3.3.2.4 Hybrid algorithm
3.3.3 Object detection algorithm
3.3.3.1 Detection algorithm based on region proposal
3.3.3.1.1 Region-based convolutional neural network
3.3.3.1.2 Spatial pyramid pooling network
3.3.3.1.3 Fast region-based convolutional neural network
3.3.3.1.4 Faster region–based convolutional neural network
3.3.3.1.5 Region-based fully convolutional networks
3.3.3.2 End-to-end detection algorithm based on deep learning
3.3.3.2.1 You Only Look Once
3.3.3.2.2 Single Shot MultiBox Detector
3.3.3.2.3 You Only Look Oncev2
3.4 Conclusion
References
Chapter-4---Identification-of-main-control-param_2021_Unmanned-Driving-Syste
4 Identification of main control parameters for train unmanned driving systems
4.1 Common methods for driving control of main control parameter identification
4.1.1 System identification
4.1.1.1 Definition of identification
4.1.1.2 Identification model
4.1.1.3 Steps of identification
4.1.1.3.1 Experimental design
4.1.1.3.2 Data processing
4.1.1.3.3 Structure identification
4.1.1.3.4 Parameter estimation
4.1.1.3.5 Model validation
4.1.1.4 Complexity, convergence, and computational efficiency of the identification algorithm
4.1.2 Common methods of parameter identification
4.1.2.1 Recursive parameter identification method
4.1.2.1.1 Recursive least square algorithm
4.1.2.1.2 Stochastic gradient algorithm
4.1.2.2 Auxiliary model identification method
4.1.2.3 Multi-innovation identification method
4.1.2.4 Iterative identification methods
4.2 Train unmanned driving dynamic models
4.2.1 Force analysis of train
4.2.1.1 Train tractive force
4.2.1.2 Train braking force
4.2.1.3 Train resistance
4.2.1.3.1 Basic resistance
4.2.1.3.2 Additional resistance
4.2.2 Dynamic model of train
4.2.2.1 Single-particle model
4.2.2.2 Multiparticle model
4.3 Identification methods of train intelligent traction
4.3.1 Fuzzy identification method
4.3.2 Simulated annealing algorithm
4.3.3 Artificial neural network
4.3.4 Genetic algorithm
4.3.5 Swarm intelligence algorithm
4.3.5.1 Ant colony optimization algorithm
4.3.5.2 Particle swarm optimization algorithm
4.3.5.3 Firefly algorithm
4.4 Conclusion
References
Chapter-5---Data-mining-and-processing-for-tr_2021_Unmanned-Driving-Systems-
5 Data mining and processing for train unmanned driving systems
5.1 Data mining and processing of manual driving modes
5.1.1 Data types of manual driving modes
5.1.2 Traditional data mining and processing technology of manual driving
5.1.2.1 Operation environment of manual driving model train
5.1.2.1.1 Line conditions
5.1.2.1.2 Train conditions
5.1.2.1.3 Other conditions
5.1.2.1.4 The process of train operation
5.1.2.2 Calculation and modeling of train traction
5.1.2.3 Strategy optimization of manual driving mode
5.1.3 Data mining and processing technology of manual driving based on the combination of offline and online
5.1.3.1 Operation environment of manual driving model train
5.1.3.2 Offline optimization of manual driving strategy based on intelligent search methods
5.1.3.3 Online optimization of manual driving strategy based on numerical iterative method
5.1.4 Data mining and processing technology of manual driving considering real-time scheduling information
5.1.4.1 Operation environment of manual driving model train
5.1.4.2 Manual driving assistance method considering real-time scheduling information
5.2 Data mining and processing of automatic driving modes
5.2.1 Data types of automatic driving modes
5.2.2 Data mining and processing technology of automatic driving based on deep learning
5.2.2.1 Operation environment of automatic driving train
5.2.2.2 Feature learning of automatic driving strategy based on deep learning
5.2.3 Data mining and processing technology of automatic driving based on adaptive differential evolution algorithm
5.2.3.1 Operation environment of automatic driving train
5.2.3.2 Strategy optimization of automatic driving based on adaptive differential evolution algorithm
5.3 Data mining and processing of unmanned driving modes
5.3.1 Data types of unmanned driving modes
5.3.2 The function of data mining technology in unmanned driving modes
5.3.2.1 Wake-up function
5.3.2.2 Dormancy function
5.3.2.3 Stop control
5.3.2.4 Emergency handling
5.3.3 Data mining and processing technology of unmanned driving modes
5.3.3.1 Background
5.3.3.2 Commonly used data mining and processing technology
5.3.3.2.1 Clustering algorithm
Partition-based clustering algorithm
Hierarchical clustering algorithm
Grid-based clustering algorithm
Density-based clustering algorithm
5.3.3.2.2 Classification algorithm
Artificial neural networks
K-nearest neighbor
Support vector machine
C4.5 decision tree
Random forest
5.3.3.2.3 Ensemble algorithm
Bagging algorithm
Boosting algorithm
AdaBoost algorithm
XGBoost algorithm
5.3.3.2.4 Machine learning algorithm
Extreme leaning machine
Backpropagation neural network
5.3.4 Comparison and analysis
5.4 Conclusion
References
Chapter-6---Energy-saving-optimization-and-contr_2021_Unmanned-Driving-Syste
6 Energy saving optimization and control for train unmanned driving systems
6.1 Technical status of train unmanned driving energy consumption analysis
6.1.1 Analysis of train operation energy consumption
6.1.2 Common train energy-saving strategies
6.1.2.1 Single train energy-saving optimization
6.1.2.2 Multiple-train collaborative optimization
6.1.2.3 Energy storage device
6.1.2.3.1 Train-mounted energy storage device
6.1.2.3.2 The ground energy storage device
6.1.3 The development and research status of energy saving optimization for train operation
6.1.3.1 Research status of single train energy saving methods
6.1.3.2 Research status of multi-train energy saving methods
6.1.3.3 Research status of energy storage device
6.1.4 Significance of optimization for train operation
6.1.4.1 Help to reduce energy consumption in the railway transport sector
6.1.4.2 An important part of the automatic train control system
6.1.4.3 There is an important theoretical significance
6.1.5 Energy consumption model of driverless train operation
6.1.5.1 Driverless train operation energy consumption model
6.1.5.2 Driverless train operation resistance model
6.2 Single-target train energy saving and manipulation based on artificial intelligence algorithm optimization
6.2.1 Optimization of energy-saving operation of driverless train based on particle swarm optimization
6.2.1.1 The theoretical basis of particle swarm optimization
6.2.1.2 Process of particle swarm optimization energy saving optimization
6.2.1.2.1 Initialization
6.2.1.2.2 The fitness value calculation of particles
6.2.1.2.3 Local optimization
6.2.1.2.4 Global optimization
6.2.1.2.5 Speed and position updating
6.2.1.2.6 Termination judgment
6.2.2 Optimization of energy-saving operation of driverless train based on the genetic algorithm
6.2.2.1 The theoretical basis of the genetic algorithm
6.2.2.1.1 Coding
6.2.2.1.2 Selection
6.2.2.1.3 Crossover
6.2.2.1.4 Mutation
6.2.2.1.5 Fitness function
6.2.2.2 Process of genetic algorithm energy saving optimization
6.3 Multiobjective train energy saving and control based on group artificial intelligence
6.3.1 Optimization of energy-saving operation of driverless train based on the multi-population genetic algorithm
6.3.1.1 The theoretical basis of the multi-population genetic algorithm
6.3.1.1.1 Principle and parameter setting of the algorithm
6.3.1.1.2 Fast nondominant sorting
6.3.1.1.3 Crowding distance
6.3.1.1.4 Elite retention strategy
6.3.1.2 Process of multi-population genetic algorithm energy saving optimization
6.3.2 Optimization of the energy saving operation of the driverless train based on the MOPSO
6.3.2.1 The theoretical basis of the MOPSO
6.3.2.1.1 Select global optimal solution and individual optimal solution
6.3.2.1.2 Establish and update external files
6.3.2.2 Process of MOPSO energy saving optimization
6.4 Conclusion
References
Chapter-7---Unmanned-driving-intelligent-algo_2021_Unmanned-Driving-Systems-
7 Unmanned driving intelligent algorithm simulation platform
7.1 Introduction of MATLAB/Simulink Simulation Platform
7.1.1 Background
7.1.2 History of train simulation software
7.1.3 MATLAB
7.1.4 Simulink
7.2 Design method of train intelligent driving algorithm simulation platform
7.2.1 Object-oriented simulation technology
7.2.2 The development process of simulation platform software
7.2.3 Description of the software architecture
7.2.4 The structure design of simulation platform software
7.3 Train automatic operation control model and programming
7.3.1 Input module
7.3.2 Controller module
7.3.3 Train model module
7.3.4 Output module
7.3.5 Basic resistance module
7.3.6 Other major modules
7.4 Train intelligent driving algorithm simulation graphical user interface design standard
7.4.1 Simulation line selection module
7.4.2 Simulation model parameter setting module
7.4.3 Algorithm selection module
7.4.4 Simulation option module
7.4.5 Display module of simulation results
7.4.5.1 Single simulation
7.4.5.2 Multiple simulation
7.5 Applications and case analysis of mainstream train unmanned driving systems
7.5.1 Principle of simulation system
7.5.2 Design of the automatic train operation algorithm
7.5.2.1 Introduction to automatic train operation algorithm
7.5.2.2 Genetic algorithms
7.5.2.3 Particle Swarm optimization
7.5.2.4 Imperial competition algorithm
7.5.2.5 Bat algorithm
7.5.2.6 Grey Wolf optimizer
7.5.2.7 Black Hole algorithm
7.5.3 Train simulation platform software testing
7.5.4 Evaluation and analysis of simulation system
7.5.4.1 Evaluation of the software
7.5.4.2 Comparison and discussion of the simulation results
7.6 Conclusion
References
Index_2021_Unmanned-Driving-Systems-for-Smart-Trains
Index