Indonesia Conference Directory

The essential problem in the estimation of a human joint angle based on electromyography (EMG) signal is the non-linearity characteristic of the EMG feature. The non-linearity of the EMG features influence the performance of the estimation. The objective of this paper is to develop an extended Kalman filter model to predict the elbow joint angle based on the EMG signal. The EMG signal is recorded from biceps muscle using disposible electrodes (Ag/AgCl). The recording of the EMG signal was conducted when the subject was instructed to perform a flexion and extension motion. In this study, a periodic and random motion were chosen to examine the proposed method. The EMG signal was extracted using sign slope feature (SSC) to obtain the information which is related to the posisition of the elbow. The response of the features was ploted to get the function of observation state. The extended Kalman filter (EKF) was chosen to linearize and to estimate the elbow joint angle based on EMG features. The performance improvement from KF to EKF based method is 12.81% and 9.65% for periodic and random motion, respectively. We have demonstrated the effectiveness of the proposed method to improve the performance of the estimation, further it can be implemented to an assistive exoskeleton for elderly people or stroke patient for better live