基于自适应切换迭代学习的绳驱踝外骨骼力控制器设计  

Force controller Design Based on Adaptively Switched Iterative Learning for Cable driven Ankle Exoskeleton

作　　者：李颖 孙容磊[2] 张歆悦 LI Ying;SUN Ronglei;ZHANG Xinyue(School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China;Institute of Medical Equipment Science and Engineering,Huazhong University of Science and Technology,Wuhan 430074,China;State Key Laboratory of Intelligent Manufacturing Equipment and Technology,Huazhong University of Science and Technology,Wuhan 430074,China)

机构地区：[1]华中科技大学机械科学与工程学院,湖北武汉430074 [2]华中科技大学医疗装备科学与工程研究院,湖北武汉430074 [3]华中科技大学智能制造装备与技术全国重点实验室,湖北武汉430074

出　　处：《机械与电子》2024年第9期36-44,共9页Machinery & Electronics

基　　金：国家自然科学基金资助项目(U21A20121,52027806)。

摘　　要：人-外骨骼构成强耦合系统,步态的偶然变异对控制器产生较大干扰,导致系统震荡。针对这一问题,提出了一种自适应切换的迭代学习控制策略。在峰值误差较大时,采用峰值迭代学习控制,仅对力控信号峰值点进行迭代更新;当峰值误差较小时,切换为传统的PD型迭代学习控制,对力控信号所有点进行迭代更新。为进一步改善控制效果,设计了前馈控制器来补偿鲍登绳传动的非线性摩擦。仿真和外骨骼实验结果表明,与传统PD型迭代学习算法相比,自适应切换的迭代学习控制策略保证了助力曲线的平稳升降,同时该策略具有更快的收敛速度以及受干扰后更快的恢复速度,实现了行走过程中辅助力的准确跟踪控制。The human exoskeleton system is a strongly coupled system,and the incidental variation in gait will cause significant disturbance to the controller,resulting in oscillations within the system.To address this issue,an adaptive switching iterative learning control strategy was proposed.When the peak error value was large,a peak iterative learning control strategy was employed,which only iteratively updated the peak point of the force control signal.When the peak error is small,the control strategy switched to the conventional PD type iterative learning control,which iteratively updated all points of the force control signal.To further improve the control effectiveness,a feedforward controller was designed to compensate for the nonlinear friction of the Bowden cable transmission.A Simulink simulation model and a flexible ankle exoskeleton experimental platform were constructed.Simulation and experimental results on the exoskeleton platform show that our peak tracking iterative learning control strategy,compared to traditional PD type algorithms,ensures smoother assistive force curve transitions.At the same time,this strategy has a faster convergence speed and a faster recovery speed after disturbances,achieving accurate tracking control of assistive force during walking process.

关 键 词：自适应切换 峰值迭代学习 PD型迭代学习 前馈补偿器 足踝外骨骼 鲍登绳传动 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]