基于MFAC-ILC的PAM-MRB柔性膝关节的仿真与试验研究  

Simulation and test study of PAM-MRB flexible knee joints based on MFAC-ILC

作　　者：罗天洪 方尚晨 马翔宇 付强 黎朝阳 LUO Tianhong;FANG Shangchen;MA Xiangyu;FU Qiang;LI Chaoyang(College of Mechanical Engineering,Chongqing University of Technology,Chongqing 400030,China;School of Intelligent Manufacturing Engineering,Chongqing University of Arts and Sciences,Chongqing 402160,China)

机构地区：[1]重庆理工大学机械工程学院,重庆400030 [2]重庆文理学院智能制造工程学院,重庆402160

出　　处：《机械传动》2025年第1期44-52,共9页Journal of Mechanical Transmission

基　　金：重庆科技局基础研究与前沿探索项目(cstc2020jcyj-msxm2322)。

摘　　要：【目的】针对气动人工肌肉(Pneumatic Artificial Muscle,PAM)与磁流变阻尼器(Magnetorheological damper,MRB)复合结构的非线性对PAM-MRB膝关节运动轨迹的跟踪精度的影响,提出一种无模型自适应控制(Model-Free Adaptive Control,MFAC)结合迭代学习控制(Iterative Learning Control,ILC)算法的方法。【方法】首先,结合气动肌肉与磁流变阻尼器的力学模型,推导出PAM-MRB柔性膝关节的动力学模型;其次,利用无模型自适应控制结合迭代算法设计出无模型自适应迭代控制器(Model-Free Adaptive Iterative Controller,MFAC-ILC),并使用Matlab软件对所提控制方法的有效性进行仿真验证;最后,搭建样机平台对提出的控制方法进行试验验证。【结果】结果表明,MFAC-ILC比比例、积分、微分(Proportion,Integral,Differential,PID)控制有着更好的控制效果,控制误差降低5°左右;相较于PID控制,MFAC-ILC对PAM-MRB柔性膝关节有着更快的轨迹跟踪收敛速度,跟踪误差降低了2°,并且避免了因为膝关节建模准确性问题带来的轨迹跟踪误差。[Objective]Aiming at the nonlinear influence of the composite structure of the pneumatic artificial muscle(PAM)and magnetorheological damper(MRB)on the tracking accuracy of the PAM-MRB knee joint motion trajectory,a modelfree adaptive control(MFAC)combined with iterative learning control(ILC)algorithm was proposed.[Methods]Firstly,combined with the mechanical model of pneumatic muscle and magnetorheological damper,the dynamic model of the PAM-MRB flexible knee joint was derived.Secondly,a model-free adaptive iterative controller(MFAC-ILC)was designed by using the model-free adaptive control combined with the iterative algorithm,and the effectiveness of the proposed control method was simulated and verified by Matlab software.Finally,a prototype platform was built to verify the proposed control method.[Results]The results show that MFAC-ILC has better control effect than the PID control,and the control error is reduced by about 5°.The MFAC-ILC has faster trajectory tracking convergence speed for the PAM-MRB flexible knee joint than the PID control,which could reduce the tracking error by 2°,and avoid the trajectory tracking error caused by the accurate knee joint modeling.

关 键 词：气动人工肌肉 磁流变阻尼器 无模型自适应控制 迭代控制 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]