东北大学学报(自然科学版) ›› 2024, Vol. 45 ›› Issue (7): 974-983.DOI: 10.12068/j.issn.1005-3026.2024.07.009

• 机械工程 • 上一篇    下一篇

上肢康复机器人模糊自适应交互控制研究

单泉(), 张顺, 黄建聪, 陈砚   

  1. 东北大学秦皇岛分校 控制工程学院,河北 秦皇岛 066004
  • 收稿日期:2023-03-15 出版日期:2024-07-15 发布日期:2024-10-29
  • 通讯作者: 单泉
  • 基金资助:
    国家自然科学基金资助项目(51905083)

Research on Fuzzy Adaptive Interactive Control of Upper Limb Rehabilitation Robots

Quan SHAN(), Shun ZHANG, Jian-cong HUANG, Yan CHEN   

  1. School of Control Engineering,Northeastern University at Qinhuangdao,Qinhuangdao 066004,China.
  • Received:2023-03-15 Online:2024-07-15 Published:2024-10-29
  • Contact: Quan SHAN
  • About author:SHAN QuanE-mail:shanquan@neuq.edu.cn

摘要:

为解决脑卒中患者在主动康复训练过程中因患者个体差异导致的训练强度不足或过强问题,提出了一种基于模糊规则的上肢康复机器人自适应交互控制系统.针对不同病情患者的肌力差异,设计模糊自适应阻抗控制器.控制器基于人机交互力和系统误差,利用模糊推理对阻尼和刚度系数进行自适应调节,改变训练强度,实现康复机器人按需辅助控制.为保证康复训练过程中运动轨迹的准确跟踪,设计GA-FuzzyPID控制器,基于改进遗传算法对模糊规则隶属度函数和规则库进行优化,降低康复机器人轨迹跟踪误差.基于Matlab/Simulink对系统进行轨迹跟踪和自适应阻抗控制仿真实验.结果表明,轨迹跟踪实验中,GA-FuzzyPID控制器的轨迹误差相较于PID控制器和FuzzyPID控制器分别降低了55.9%和34.0%,有效提高了轨迹跟踪精度;自适应阻抗控制实验通过与固定阻抗方法进行对比,验证了所提自适应交互控制系统的有效性和可行性.

关键词: 上肢康复机器人, 阻抗控制, 模糊规则, 遗传算法, 自适应控制

Abstract:

An adaptive interactive control system based on fuzzy rules for upper limb rehabilitation robots is proposed to address the insufficient or excessive training intensity during active rehabilitation exercises for stroke patients due to their individual differences. According to the difference of muscle strength of patients with different conditions, a fuzzy adaptive impedance controller is designed, which adjusts the damping and stiffness coefficients adaptively with fuzzy inference based on human?machine interaction forces and system errors, altering the training intensity to achieve on?demand assistive control for rehabilitation robots. Additionally, to ensure accurate tracking of the motion trajectory during rehabilitation training, a GA-FuzzyPID controller is designed to optimize the fuzzy rule membership functions and rule base according to an improved genetic algorithm, thereby reducing the trajectory tracking error of rehabilitation robots. Finally, trajectory tracking and adaptive impedance controlling simulation experiments are conducted for the system based on Matlab/Simulink. The results show that in the trajectory tracking experiment, the trajectory error of GA-FuzzyPID controller is reduced by 55.9% and 34.0% respectively compared with PID controller and FuzzyPID controller, which effectively improves the trajectory tracking accuracy. Compared with the fixed impedance method, the adaptive impedance control experiment verifies the effectiveness and feasibility of the proposed adaptive interactive control system.

Key words: upper limb rehabilitation robot, impedance control, fuzzy rule, genetic algorithm, adaptive control

中图分类号: