A method of discriminating fingers and wrist action from surface EMG signals for controlling robotic or prosthetic forearm hand

Abstract

The purpose of the study is to find a method for discriminating wrist action and finger action from electro-miography (EMG) picked up from surface electrodes. It will contribute a voluntary control of robotic or prosthetic forearm hand that has wrist joint and fingers. The proposed method is based on two physiological facts: the first is that EMG coming from deep layer has a tendency of power spectrum shifting to low frequency: the second is that EMG coming from thin layer shows a difference in the magnitude according to difference in distance between differential electrodes. The double differential electrode with pre-amplifiers that has two pair of electrodes with different electrode distances is then developed to confirm the method. Experiments show the proposed method is capable of discriminating finger action, wrist action and their combinatory action.