Optimal measurement position estimation of EMG signal by multi-regression analysis for human forearm motion discrimination

This paper describes an optimal measurement position estimation of the myoelectric signal by the multiple regression analysis for human forearm motion discrimination. The conventional studies use a lot of myoelectric sensors to obtain the high discrimination precision from the myoelectric signal. However, the use of a lot of myoelectric sensors becomes difficult by the cost and amputating situation of the human forearm. The purpose of this study is to decide the optimal measurement position for the motion discrimination, and to obtain high discrimination precision of the human forearm motion. This study proposes the selection method of the optimal measurement position to estimate the identification target motion from the myoelectric signal measured from multiple positions by the multiple regression analysis. Some experiments on the myoelectric hand simulator show the effectiveness of the proposed optimal measurement position selection method.