Scholarlink

<p>To address the challenges of asymmetric time-varying delays and dynamic model uncertainties in bilateral teleoperation systems, this paper proposes an adaptive finite-time control strategy integrating zeroing neural dynamics (ZND) optimization with interval type-2 fuzzy neural network (IT2FNN). Specifically, embedding the ZND optimization method defines idealized convergence dynamics for composite error variables. Unlike conventional passive error control approaches, this method actively generates a reference acceleration to guide error convergence, thereby enhancing error convergence accuracy. Concurrently, the IT2FNN leverages its uncertainties-handling capability to perform online approximation of the system dynamic model, ensuring stability under time-delay conditions and resolving dependency on physically modeled drivers. Furthermore, finite-time compensation terms are designed to achieve rapid error convergence. Stability analysis conducted via the Lyapunov–Krasovskii method verifies the stability of the closed-loop teleoperation system. Finally, comparative experiments validate that the proposed scheme achieves higher trajectory tracking accuracy than traditional adaptive control methods, fuzzy control methods and radial basis function neural networks.</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

Zeroing Neural Dynamics-Based Adaptive Interval Type-2 Fuzzy Neural Network Control for Bilateral Teleoperation Systems

  • 期刊论文

  • 发表日期：

    2026年5月11日

查看全文
  • Erchao Li,
  • Ze Wang,
  • Haochen Zhang

摘要

To address the challenges of asymmetric time-varying delays and dynamic model uncertainties in bilateral teleoperation systems, this paper proposes an adaptive finite-time control strategy integrating zeroing neural dynamics (ZND) optimization with interval type-2 fuzzy neural network (IT2FNN). Specifically, embedding the ZND optimization method defines idealized convergence dynamics for composite error variables. Unlike conventional passive error control approaches, this method actively generates a reference acceleration to guide error convergence, thereby enhancing error convergence accuracy. Concurrently, the IT2FNN leverages its uncertainties-handling capability to perform online approximation of the system dynamic model, ensuring stability under time-delay conditions and resolving dependency on physically modeled drivers. Furthermore, finite-time compensation terms are designed to achieve rapid error convergence. Stability analysis conducted via the Lyapunov–Krasovskii method verifies the stability of the closed-loop teleoperation system. Finally, comparative experiments validate that the proposed scheme achieves higher trajectory tracking accuracy than traditional adaptive control methods, fuzzy control methods and radial basis function neural networks.