Scholarlink

<p>Aiming at the problem of chattering and trajectory tracking performance degradation caused by system uncertainty, modeling errors, and unknown gap-like time delay in a multi-joint manipulator, an adaptive neural network control approach utilizing an enhanced hybrid algorithm is proposed. First, considering the unknown dynamics in the manipulator model, a radial basis function neural network (RBFNN) is employed for approximation, and the RBFNN parameters are optimized using chaotic mapping and an improved hybrid algorithm to enhance its approximation capability. Second, to compensate for the errors caused by unknown gap-like time delays and RBFNN approximation, an adaptive reaching law is proposed, which not only improves control performance but also suppresses chattering. Finally, a 3-DOF robotic arm is used as the control object, and comparative simulations are conducted to verify the robustness and superiority of the proposed control scheme.</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

Adaptive sliding mode control of gap-like manipulators based on an improved hybrid algorithm-optimized neural network

  • 期刊论文

  • 发表日期：

    2025年11月8日

查看全文
  • Jiqing Chen,
  • Bin Lu,
  • Mingchang Zhang,
  • Wei Wang,
  • Peilin Li

摘要

Aiming at the problem of chattering and trajectory tracking performance degradation caused by system uncertainty, modeling errors, and unknown gap-like time delay in a multi-joint manipulator, an adaptive neural network control approach utilizing an enhanced hybrid algorithm is proposed. First, considering the unknown dynamics in the manipulator model, a radial basis function neural network (RBFNN) is employed for approximation, and the RBFNN parameters are optimized using chaotic mapping and an improved hybrid algorithm to enhance its approximation capability. Second, to compensate for the errors caused by unknown gap-like time delays and RBFNN approximation, an adaptive reaching law is proposed, which not only improves control performance but also suppresses chattering. Finally, a 3-DOF robotic arm is used as the control object, and comparative simulations are conducted to verify the robustness and superiority of the proposed control scheme.