基于sEMG的下肢运动解析方法研究  

Research on lower limb kinematic analysis method based on sEMG

作　　者：邱石 杜义浩 王浩 谢平 于金须 Qiu Shi;Du Yihao;Wang Hao;Xie Ping;Yu Jinxu(Key Laboratory of Measurement Technology and Instrumentation of Hebei Province, Institute of Electric Engineering, Yanshan University, Qinhuangdao 066004, Chin)

机构地区：[1]燕山大学电气工程学院河北省测试计量技术及仪器重点实验室

出　　处：《仪器仪表学报》2018年第2期30-37,共8页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61673336,61503325);中国博士后科学基金(2015M581316);河北省教育厅高等学校科技计划(QN2016094)项目资助

摘　　要：为了实现肢体运动解析并用于康复机器人系统自适应控制,提出一种基于sEMG的下肢运动解析方法。同步采集下肢6块肌肉的表面肌电信号（sEMG）和髋、膝关节角度;引入相干性分析方法,定量描述s EMG和关节角度耦合关系,进而优化选取肌肉通道;采用一阶递归滤波器补偿sEMG和关节角度的机电延迟（EMD）;提出基于黄金分割的最小二乘极限学习机（GSLSELM）算法,进行下肢运动解析。实验结果表明,7名被试下肢运动解析的均方根误差（RMSE）和时长,能够满足康复机器人系统控制的准确性和实时性要求。In order to realize limb motion analysis and use it in the adaptive control for rehabilitation robot system,a new lower limb motion analysis method is proposed based on sEMG signal in this paper. The surface Electromyography（ sEMG） signals of 6 lower limb muscles and the hip,knee joint angles are acquired synchronously; Coherent analysis method is introduced and used to describe the coupling relationship between s EMGs and joint angles quantitatively,and then the optimal muscle channels are selected. A first-order recursive filter is used to compensate the electromechanical delay between the s EMGs and the joint angles. The least square extreme learning machine algorithm based on golden section（ GS-LSELM） is proposed to analyze the lower limb motion information. The experiment results show that the proposed method achieves smaller RMSE and shorter computation time for the lower limb motion analysis of 7 subjects. Consequently,the method can meet the real-time and high accuracy requirements of the rehabilitation robot system control.

关 键 词：表面肌电信号 相干性分析 肌肉筛选 基于黄金分割的最小二乘极限学习机 运动解析 

分 类 号：TH911.72[机械工程]