SEMG signal and hill model based continuous prediction for hand grasping motion

This paper is aimed at the continuous hand grasping motion prediction during all fingers flexion and extension. Only sEMG signals recorded from flexor digitorum superficialis and extensor digitorum of forearm are used to predict the flexion and extension motion. In order to find the relation between sEMG signals and hand grasping motion, a Hill model is used to represent the force value of the muscles. Some assumptions are also made for simplicity in calculating the association. A simple and efficient motion recording system using flex sensor, Mtx sensor and a glove is designed for the purpose of recording fingers motion. The motions are voluntary finger flexion and extension with no load. Acceptable results are achieved. The purpose of this paper is to provide a method for continuous hand grasping motion prediction based on sEMG signals. Although some assumptions are made to simplify the problem and indeed these assumptions brought prediction errors in the experiment, the method shows itself an alternative way to use sEMG signals for hand motion prediction.