Scholarlink

This paper presents a robust neural adaptive output-feedback control strategy for multi-input-multi-output (MIMO) systems with time-varying delays. A linear state observer addresses unavailable state variables, while a neural network approximates unknown nonlinear functions. The control law mitigates external disturbances and various errors. The Gradient Algorithm with Projection tackles unknown control directions, and the Strictly Positive Real (SPR) condition, applied through the Lyapunov–Krasovskii method, aids in designing adaptation laws using output errors. This approach offers several advantages: it applies to a wide range of MIMO systems, avoids singularity issues, requires few adapting parameters, and ensures asymptotic convergence of tracking errors. A simulation example validates the method's effectiveness and feasibility.

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 Network-Based Robust Adaptive Output Feedback Control for MIMO Time-Varying Delay Systems

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Farouk Zouari,
  • Mufti Mahmud

摘要

This paper presents a robust neural adaptive output-feedback control strategy for multi-input-multi-output (MIMO) systems with time-varying delays. A linear state observer addresses unavailable state variables, while a neural network approximates unknown nonlinear functions. The control law mitigates external disturbances and various errors. The Gradient Algorithm with Projection tackles unknown control directions, and the Strictly Positive Real (SPR) condition, applied through the Lyapunov–Krasovskii method, aids in designing adaptation laws using output errors. This approach offers several advantages: it applies to a wide range of MIMO systems, avoids singularity issues, requires few adapting parameters, and ensures asymptotic convergence of tracking errors. A simulation example validates the method's effectiveness and feasibility.