Reinforcement learning-based MPC for tracking control of 4WID4WIS

被引：1

作者：

Lin, Wenlong ^{[1
]}

Wang, Ping ^{[1
]}

Wu, Yu-Yao ^{[1
]}

Liu, Wanquan ^{[1
]}

Sun, Hui-Jie ^{[1
]}

机构：

[1] Sun Yat Sen Univ, Shenzhen Campus, Shenzhen 518107, Guangdong, Peoples R China

来源：

2023 2ND CONFERENCE ON FULLY ACTUATED SYSTEM THEORY AND APPLICATIONS, CFASTA | 2023年

基金：

中国国家自然科学基金;

关键词：

Reinforcement learning; Model predictive control; Trajectory tracking; 4WID4WIS; MODEL-PREDICTIVE CONTROL;

D O I：

10.1109/CFASTA57821.2023.10243247

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

To achieve trajectory tracking with 4wiswid(Four wheel independent drive and Four wheel-independent steering), a reinforcement learning-based linear model predictive control(MPC) controller is designed. The controller is designed based on the kinematics model of 4WID4WIS, but the dynamics are unknown. Adding reinforcement learning to learn errors caused by unknown dynamics, adjust the weight matrix to reduce the error, and improve the control performance of the system. The stability of the proposed reinforcement learning-based MPC has been demonstrated by Lyapunov. The proposed control method has a fast response speed and good robustness. Simulation and experiments are given to demonstrate the availability of this method.

引用

页码：839 / 844

页数：6

共 21 条

[1]

Bemporad A, 1999, LECT NOTES CONTR INF, V245, P207

[2] Min-max control of constrained uncertain discrete-time linear systems [J].

Bemporad, A ;

Borrelli, F ;

Morari, M .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2003, 48 (09) :1600-1606

[3] Adaptive Partial Reinforcement Learning Neural Network-Based Tracking Control for Wheeled Mobile Robotic Systems [J].

Ding, Liang ;

Li, Shu ;

Gao, Haibo ;

Chen, Chao ;

Deng, Zongquan .

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2020, 50 (07) :2512-2523

[4]

Esfahani HN, 2021, 2021 EUROPEAN CONTROL CONFERENCE (ECC), P132, DOI 10.23919/ECC54610.2021.9655129

[5] Getting Robustness Against Unstructured Uncertainty: A Tube-Based MPC Approach [J].

Falugi, Paola ;

Mayne, David Q. .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2014, 59 (05) :1290-1295

[6] An Intelligent Non-Integer PID Controller-Based Deep Reinforcement Learning: Implementation and Experimental Results [J].

Gheisarnejad, Meysam ;

Khooban, Mohammad Hassan .

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2021, 68 (04) :3609-3618

[7] Design of an Active Collision Avoidance System for a 4WIS-4WID Electric Vehicle [J].

Hang, Peng ;

Han, Yanqun ;

Chen, Xinbo ;

Zhang, Bang .

IFAC PAPERSONLINE, 2018, 51 (31) :771-777

[8]

Hang P, 2017, CHIN CONTR CONF, P9360, DOI 10.23919/ChiCC.2017.8028849

[9]

Kim Y., 2022, 2022 INT C EL INF CO, P1

[10] Actor-Critic Learning Control With Regularization and Feature Selection in Policy Gradient Estimation [J].

Li, Luntong ;

Li, Dazi ;

Song, Tianheng ;

Xu, Xin .

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2021, 32 (03) :1217-1227

← 1 2 3 →