Scholarlink

<p>This paper addresses the joint vibration and external disturbance problems encountered by manipulators in agricultural inspection robots during intelligent cultivation processes. A novel three-layer hybrid vibration suppression architecture is proposed, integrating sliding mode observer (SMO) enhanced model predictive control (MPC) with zero vibration and derivative (ZVD) input shaping. First, a 12th-order dynamic model of a three-link flexible joint manipulator is established, and the multi-modal vibration characteristics along with disturbance sources are systematically analyzed. Second, a sliding mode observer based on the super-twisting algorithm is designed for disturbance estimation and state reconstruction, an MPC controller with integrated disturbance compensation is developed, and a ZVD input shaper based on the lowest natural frequency is designed. Third, the asymptotic stability of the closed-loop system is rigorously proved using Lyapunov stability theory. Fourth, comparative analysis with active disturbance rejection control (ADRC) methods is provided to demonstrate the advantages of the proposed approach. Finally, comprehensive simulation experiments are conducted in the MATLAB/Simulink environment. Results demonstrate that compared with the conventional MPC-ZVD method, the proposed SMO-MPC-ZVD enhanced architecture reduces peak vibration by 42.3%, residual vibration RMS by 58.7%, and settling time by 31.2%. The method exhibits an average single-step computation time of 0.42&#xa0;ms, providing sufficient computational margin for real-time control at 1&#xa0;kHz sampling rate. Excellent control performance is maintained under ± 25% parameter uncertainty and step disturbance conditions, with residual vibration remaining below 0.02°. This method provides an effective technical solution for precise motion control of facility agriculture inspection robots in complex environments.</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

Vibration suppression of flexible joint manipulator for agricultural inspection robot based on SMO-MPC-ZVD enhanced architecture

  • 期刊论文

  • 发表日期：

    2026年5月4日

查看全文
  • Hao Tan,
  • Shuang Zhao,
  • Yubin Miao

摘要

This paper addresses the joint vibration and external disturbance problems encountered by manipulators in agricultural inspection robots during intelligent cultivation processes. A novel three-layer hybrid vibration suppression architecture is proposed, integrating sliding mode observer (SMO) enhanced model predictive control (MPC) with zero vibration and derivative (ZVD) input shaping. First, a 12th-order dynamic model of a three-link flexible joint manipulator is established, and the multi-modal vibration characteristics along with disturbance sources are systematically analyzed. Second, a sliding mode observer based on the super-twisting algorithm is designed for disturbance estimation and state reconstruction, an MPC controller with integrated disturbance compensation is developed, and a ZVD input shaper based on the lowest natural frequency is designed. Third, the asymptotic stability of the closed-loop system is rigorously proved using Lyapunov stability theory. Fourth, comparative analysis with active disturbance rejection control (ADRC) methods is provided to demonstrate the advantages of the proposed approach. Finally, comprehensive simulation experiments are conducted in the MATLAB/Simulink environment. Results demonstrate that compared with the conventional MPC-ZVD method, the proposed SMO-MPC-ZVD enhanced architecture reduces peak vibration by 42.3%, residual vibration RMS by 58.7%, and settling time by 31.2%. The method exhibits an average single-step computation time of 0.42 ms, providing sufficient computational margin for real-time control at 1 kHz sampling rate. Excellent control performance is maintained under ± 25% parameter uncertainty and step disturbance conditions, with residual vibration remaining below 0.02°. This method provides an effective technical solution for precise motion control of facility agriculture inspection robots in complex environments.