Optimal measurement position estimation by discriminant analysis based on Wilks´ lambda for myoelectric hand control

This paper describes an optimal measurement position estimation by the discriminant analysis based on Wilks´ lambda for the myoelectric hand control. In the past studies, the myoelectric signals were measured from the same positions for the motions discrimination. However, the optimal measurement positions of the myoelectric signals for the motion discrimination are different according to the remaining muscle situation of amputees. Therefore the purpose of this study is to estimate the optimal and fewer measurement positions for the precise motion discrimination of the human forearm. This study proposes the estimation method of the optimal measurement positions by the discriminant analysis based on Wilks´ lambda among the myoelectric signal measured from multiple positions. Some experiments on the myoelectric hand simulator show the effectiveness of the proposed optimal measurement position estimation method.