System for the Identification of Multiple Movements of the Hand

Abstract

The principal objective of this work is to develop a system of myoelectric control for the teleoperation of a complex robotic arm. It is presented a scheme for the identification of 7 movements of the hand, taking into account 3 anatomical positions of the subject and using only 2 electromyography (EMG) channels. When presenting an input to the system, this one generates an answer codified in binary of the movement the operator is making. The obtained results demonstrate that it is feasible the identification of multiple movements of the hand by using of the bioelectric signal produced by muscles of the forearm and the use of artificial neural networks (ANN) like as the classification element