基于sEMG的多自由度下肢外骨骼康复机器人结构与控制策略  

Structural and Control Strategy of Multi-DOF Lower Limb Exoskeleton Rehabilitation Robot Based on sEMG

作　　者：陈龙 梁辉[1,2] 王辉 矫恒安 汪传生[1,2] CHEN Long;LIANG Hui;WANG Hui;JIAO Heng'an;WANG Chuansheng(College of Electromechanical Engineering,Qingdao University of Science and Technology,Qingdao 266061,China;National Engineering Laboratory for Advanced Equipments and Key Materials for Tires,Qingdao University of Science and Technology,Qingdao 266061,China)

机构地区：[1]青岛科技大学机电工程学院,山东青岛266061 [2]青岛科技大学轮胎先进装备与关键材料国家工程研究中心,山东青岛266061

出　　处：《青岛科技大学学报（自然科学版）》2024年第4期121-128,共8页Journal of Qingdao University of Science and Technology:Natural Science Edition

基　　金：国家自然科学基金项目(52173101).

摘　　要：提出一种多自由度下肢外骨骼康复机器人,能够实现下肢多种康复训练。建立踝关节部分约束方程,采用数值法证明机构存在正确解。利用Delsys设备提取脚部运动的sEMG信号,采用LDA、RNN结合LSTM、CNN 3种信号分类方法,提出降维CNN方法,对输入运动进行识别和分类。最后,针对踝关节部分进行sEMG信号作为输出指令的脚部动作反馈实验,验证了该部分机构和控制方法的可行性和合理可靠性。A multi-DOF lower-limb exoskeleton rehabilitation robot was designed to realize the rehabilitation training of the lower limbs.The constraint equation of the ankle partial was established,and the numerical method was used to prove that the mechanism has a correct solution.Then the sEMG signals of motions of the feet are extracted by Delsys equipment.The data is classified by the three classification algorithms of LDA,RNN combined with LSTM and CNN,a dimensionality reduction CNN method is proposed,the input motion is identified and classified.Finally,a foot action feedback experiment with the sEMG signal as the output command was carried out for the ankle joint part to verify the feasibility and reasonable reliability of the mechanism and control method of this part.

关 键 词：外骨骼康复机器人 表面肌电信号 分类识别 运动控制 

分 类 号：TH112[机械工程—机械设计及理论] R496[医药卫生—康复医学]