Scholarlink

<p>This study proposes a fast fixed-time sliding mode trajectory tracking control strategy for autonomous surface vehicles (ASVs) under unknown dynamics and environmental disturbances. Three key innovations distinguish our work: 1) A novel fast fixed-time stable system is developed, achieving faster convergence compared to existing methods through a dual-phase state acceleration mechanism and gain scheduling function; 2) Unlike most disturbance observers, which typically offer a single estimation capability, we propose a neural network-based fixed-time disturbance observer that simultaneously approximates unknown ASV dynamics and reconstructs compound disturbances without any prior knowledge, thereby providing more accurate compensation for system uncertainties; 3) A nonsingular fixed-time sliding mode tracking controller is developed, which incorporates a newly developed sliding mode surface and an arctangent function to ensure fast error convergence and overcome control singularities. The fixed-time stability of the closed-loop system is established using Lyapunov theory. Finally, simulation results are presented on an ASV to validate the proposed theoretical findings.</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

Neural Disturbance Observer-based Fast Fixed-time Sliding Mode Tracking Control for Autonomous Surface Vehicles With Uncertain Dynamics

  • 期刊论文

  • 发表日期：

    2025年7月14日

查看全文
  • Taiqi Wang,
  • Yongtao Liu,
  • Shifeng Niu,
  • Xinfeng Zhang

摘要

This study proposes a fast fixed-time sliding mode trajectory tracking control strategy for autonomous surface vehicles (ASVs) under unknown dynamics and environmental disturbances. Three key innovations distinguish our work: 1) A novel fast fixed-time stable system is developed, achieving faster convergence compared to existing methods through a dual-phase state acceleration mechanism and gain scheduling function; 2) Unlike most disturbance observers, which typically offer a single estimation capability, we propose a neural network-based fixed-time disturbance observer that simultaneously approximates unknown ASV dynamics and reconstructs compound disturbances without any prior knowledge, thereby providing more accurate compensation for system uncertainties; 3) A nonsingular fixed-time sliding mode tracking controller is developed, which incorporates a newly developed sliding mode surface and an arctangent function to ensure fast error convergence and overcome control singularities. The fixed-time stability of the closed-loop system is established using Lyapunov theory. Finally, simulation results are presented on an ASV to validate the proposed theoretical findings.