Abstract

Info

Title:: Prediction of knee joint angle when crossing obstacles based on myoelectric signals

Author(s):: CHEN Tianlin1; DAI Quanmin1; 2; CHENG Guang1; 2; MA Yongjie1; SUN Baixin1; LIU Weifeng1; XU Xiaorong2; 1. College of Robotics, Beijing Union University, Beijing 100101, China 2. College of Urban Rail Transit and Logistics, Beijing Union University, Beijing 100101, China

Keywords:: Keywords: knee joint obstacle crossing myoelectrical signal BP neural network angle prediction

Abstract:: Abstract: In order to solve the problem of knee joint angle output when the human body crosses obstacles, a wearable signal acquisition test bench is designed. The motion analysis of the lower limb is carried out, and the myoelectrical signals and joint angle signals are used as motion data. After signal processing, BP neural network is used to predict the output angle when crossing obstacles. Herein a novel method based on BP neural network algorithm is proposed to analyze the forces of knee joint motion active muscle and passive muscle according to different thigh lift heights, and to predict the knee joint angle when the human body crosses obstacles. The proposed method can effectively help the prosthetic knee joint or rehabilitation robots implement the complex movement of obstacle crossing.