SURFACE EMG FOR MULTI-PATTERN RECOGNITION WITH SENSORY FEEDBACK CONTROLLER OF HAND PROSTHESIS SYSTEM

摘要

A bio-inspired controller have been developed based on the HIT/DLR Prosthetic Hand II in order to help the amputees perceive and use the prosthetic hands more like people with normal physiological hands. The controller includes two strategies for hand control. Srategy I has been designed to interpret the amputee's intensions through electromyography (EMG) signals recognition. It can recognize 10 types of hand postures through six electrodes by using support vector machine (SVM) classifier. In strategy II, EMG controller, vibrotactile system, torque sensors and motor drivers of the hand constructed an anthropomorphic close loop. The vibrotactile system can provide the subject with grasping force sense. The closed-loop system has established the subject-prosthesis interface control, reduce the hand mis-manipulation and force delivery mistakes. These strategies and experiments pave the way for a better control of advanced prosthetic hand and open up a scenario in which life conditions of amputees can be effectively improved.