Reinforcement learning based parameter optimization of active disturbance rejection control for autonomous underwater vehicle

• 作者: SONG Wanping CHEN Zengqiang SUN Mingwei SUN Qinglin
• 来源期刊: 系统工程与电子技术（英文版）
• 年，卷（期）: 2022,（1）
• 所属期刊栏目: CONTROL THEORY AND APPLICATION
• 页码范围: 170-179
• 页数: 10页
• 语种: 英文
• DOI: 10.23919/JSEE.2022.000017

摘要

This paper proposes a liner active disturbance rejec-tion control(LADRC)method based on the Q-Learning al-gorithm of reinforcement learning(RL)to control the six-degree-of-freedom motion of an autonomous underwater vehicle(AUV).The number of controllers is increased to realize AUV motion de-coupling.At the same time,in order to avoid the oversize of the algorithm,combined with the controlled content,a simplified Q-learning algorithm is constructed to realize the parameter adapt-ation of the LADRC controller.Finally,through the simulation ex-periment of the controller with fixed parameters and the control-ler based on the Q-learning algorithm,the rationality of the sim-plified algorithm,the effectiveness of parameter adaptation,and the unique advantages of the LADRC controller are verified.