Scholarlink

<p>To improve the response speed, working accuracy and anti-interference ability of collaborative robots, a collaborative robot control method based on model prediction and adaptive algorithm is proposed. A non-parametric model predictive current control method is employed to update current data in real time. It can effectively reduce the impact of parameter variations on control performance. Meanwhile, at the linkage end, a nonmodel-based adaptive control is adopted and combined with the recursive least squares (RLS) algorithm for online dynamic parameter estimation. This method eliminates the need for extra sensors and thus reducing costs. By integrating alternative system functions with a nonmodel-based adaptive algorithm to estimate their outputs, this method diverges from traditional estimation techniques based on the original dynamics model, thereby circumventing the need for measuring angular acceleration and integration operations and avoiding error accumulation. Experimental results demonstrate that the proposed method not only enhances the system’s dynamic response speed and tracking accuracy but also significantly improves its robustness in the presence of complex external disturbances.</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

Robustness Enhancement Strategy for Flexible Joints Based on Model Predictive Current and Adaptive Control

  • 期刊论文

  • 发表日期：

    2025年9月24日

查看全文
  • Yifei Yang,
  • Jianxing Zhao,
  • Xiaodong Sun,
  • Yong Wang

摘要

To improve the response speed, working accuracy and anti-interference ability of collaborative robots, a collaborative robot control method based on model prediction and adaptive algorithm is proposed. A non-parametric model predictive current control method is employed to update current data in real time. It can effectively reduce the impact of parameter variations on control performance. Meanwhile, at the linkage end, a nonmodel-based adaptive control is adopted and combined with the recursive least squares (RLS) algorithm for online dynamic parameter estimation. This method eliminates the need for extra sensors and thus reducing costs. By integrating alternative system functions with a nonmodel-based adaptive algorithm to estimate their outputs, this method diverges from traditional estimation techniques based on the original dynamics model, thereby circumventing the need for measuring angular acceleration and integration operations and avoiding error accumulation. Experimental results demonstrate that the proposed method not only enhances the system’s dynamic response speed and tracking accuracy but also significantly improves its robustness in the presence of complex external disturbances.