Design and Implementation of Deep Neural Network-Based Control for Automatic Parking Maneuver Process

被引:177
作者
Chai, Runqi [1 ]
Tsourdos, Antonios [1 ]
Savvaris, Al [1 ]
Chai, Senchun [2 ]
Xia, Yuanqing [2 ]
Chen, C. L. Philip [3 ,4 ,5 ]
机构
[1] Cranfield Univ, Sch Aerosp Transport & Mfg, Bedford MK43 OAL, England
[2] Beijing Inst Technol, Sch Automat, Beijing 100081, Peoples R China
[3] South China Univ Technol, Sch Comp Sci & Engn, Guangzhou 510006, Peoples R China
[4] Dalian Maritime Univ, Nav Coll, Dalian 116026, Peoples R China
[5] Univ Macau, Fac Sci & Technol, Taipa, Macao, Peoples R China
关键词
Trajectory optimization; Computational modeling; Roads; Real-time systems; Land vehicles; Autonomous vehicles; Artificial neural networks; Autonomous ground vehicles (AGVs); deep neural network (DNN); motion controller; parking maneuver; trajectory optimization; TRAJECTORY OPTIMIZATION; AUTONOMOUS VEHICLES; TRACKING CONTROL;
D O I
10.1109/TNNLS.2020.3042120
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This article focuses on the design, test, and validation of a deep neural network (DNN)-based control scheme capable of predicting optimal motion commands for autonomous ground vehicles (AGVs) during the parking maneuver process. The proposed design utilizes a multilayer structure. In the first layer, a desensitized trajectory optimization method is iteratively performed to establish a set of time-optimal parking trajectories with the consideration of noise-perturbed initial configurations. Subsequently, by using the preplanned optimal parking trajectory data set, several DNNs are trained in order to learn the functional relationship between the system state-control actions in the second layer. To obtain further improvements regarding the DNN performances, a simple yet effective data aggregation approach is designed and applied. These trained DNNs are then utilized as the motion controllers to generate feedback actions in real time. Numerical results were executed to demonstrate the effectiveness and the real-time applicability of using the proposed control scheme to plan and steer the AGV parking maneuver. Experimental results were also provided to justify the algorithm performance in real-world implementations.
引用
收藏
页码:1400 / 1413
页数:14
相关论文
共 33 条
[1]   Neural-Network-Based Adaptive Backstepping Control With Application to Spacecraft Attitude Regulation [J].
Cao, Xibin ;
Shi, Peng ;
Li, Zhuoshi ;
Liu, Ming .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (09) :4303-4313
[2]   Real-Time Reentry Trajectory Planning of Hypersonic Vehicles: A Two-Step Strategy Incorporating Fuzzy Multiobjective Transcription and Deep Neural Network [J].
Chai, Runqi ;
Tsourdos, Antonios ;
Savvaris, Al ;
Xia, Yuanqing ;
Chai, Senchun .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2020, 67 (08) :6904-6915
[3]   Two-Stage Trajectory Optimization for Autonomous Ground Vehicles Parking Maneuver [J].
Chai, Runqi ;
Tsourdos, Antonios ;
Savvaris, Al ;
Chai, Senchun ;
Xia, Yuanqing .
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2019, 15 (07) :3899-3909
[4]   Trajectory Optimization of Space Maneuver Vehicle Using a Hybrid Optimal Control Solver [J].
Chai, Runqi ;
Al Savvaris ;
Tsourdos, Antonios ;
Chai, Senchun ;
Xia, Yuanqing .
IEEE TRANSACTIONS ON CYBERNETICS, 2019, 49 (02) :467-480
[5]   Constrained Global Path Optimization for Articulated Steering Vehicles [J].
Choi, Ji-wung ;
Huhtala, Kalevi .
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, 2016, 65 (04) :1868-1879
[6]   A Survey of Methods Available for the Numerical Optimization of Continuous Dynamic Systems [J].
Conway, Bruce A. .
JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 2012, 152 (02) :271-306
[7]   From Reeds and Shepp's to continuous-curvature paths [J].
Fraichard, T ;
Scheuer, A .
IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION, 2004, 20 (06) :1025-1035
[8]   Maneuver-Based Trajectory Planning for Highly Autonomous Vehicles on Real Road With Traffic and Driver Interaction [J].
Glaser, Sebastien ;
Vanholme, Benoit ;
Mammar, Said ;
Gruyer, Dominique ;
Nouveliere, Lydie .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2010, 11 (03) :589-606
[9]   A survey of deep learning techniques for autonomous driving [J].
Grigorescu, Sorin ;
Trasnea, Bogdan ;
Cocias, Tiberiu ;
Macesanu, Gigel .
JOURNAL OF FIELD ROBOTICS, 2020, 37 (03) :362-386
[10]   Adaptive Fault-Tolerant Spacecraft Pose Tracking With Control Allocation [J].
Gui, Haichao ;
de Ruiter, Anton H. J. .
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, 2019, 27 (02) :479-494