Scholarlink

<p>In this paper, an optimized backstepping (OB) with disturbance observer control problem is explored for unmanned surface vessels (USVs) course tracking system. Firstly, in order to degrade the influence of external disturbance caused by wind, waves and current, a disturbance observer is applied in estimating the external disturbance and compensates for the controller, which does not require the prior knowledge of the complex marine environment. Secondly, an OB method by combining reinforcement learning (RL) with backstepping technique is developed for USVs course tracking control system. Different from other optimal methods, the OB method is to design the controller of every backstepping step to be the optimized solution of the corresponding subsystem so that the overall system control is optimized. Finally, the stability of the USVs course tracking control system is guaranteed and all signals in the closed-loop system are uniformly ultimately bounded (UUB) on the basis of the Lyapunov theory. Simulation results validate the effectiveness of the proposed method.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Reinforcement Learning-Based Optimal Control for USVs Course Tracking under Disturbances

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月28日

查看全文
  • Peng Li,
  • Weiwei Bai,
  • Wenjun Zhang,
  • Yikai Xv

摘要

In this paper, an optimized backstepping (OB) with disturbance observer control problem is explored for unmanned surface vessels (USVs) course tracking system. Firstly, in order to degrade the influence of external disturbance caused by wind, waves and current, a disturbance observer is applied in estimating the external disturbance and compensates for the controller, which does not require the prior knowledge of the complex marine environment. Secondly, an OB method by combining reinforcement learning (RL) with backstepping technique is developed for USVs course tracking control system. Different from other optimal methods, the OB method is to design the controller of every backstepping step to be the optimized solution of the corresponding subsystem so that the overall system control is optimized. Finally, the stability of the USVs course tracking control system is guaranteed and all signals in the closed-loop system are uniformly ultimately bounded (UUB) on the basis of the Lyapunov theory. Simulation results validate the effectiveness of the proposed method.