Design And Myoelectrical Control Of Anthropomorphic Prosthetic Hand

The myoelectric hand is a type of anthropomorphic robotic hand that driven by the human electromyography(EMG) signals for amputee rehabilitation. It is aimed at providing the amputee rehabilitation with a prosthetic hand with light weight, small size and most grasping and operating functions of human hands. The study is not only of great theory and market values but also plays an important role in improving the abilities of amputee rehabilitation and the progress of the whole society. This paper presents a design of an anthropomorphic prosthetic hand. The hand has 5 fingers and 4 degrees of freedom(DOFs) driven by 4 independent actuators. Helical springs are used as elastic joints and the joints of each finger are coupled by tendons. The compact drive system is settled in the palm to control the fingers. The simulation of the prosthetic hand show that the hand can execute most grasp tasks.The prosthetic hand is controlled by the surface electromyography(sEMG) signals from 2 electrodes only. The myoelectric control system which can classify 8 prehensile hand gestures has been built. It consists of 4 basic parts, signal source, data collection, data processing and prosthetic hand control. Pattern recognition is employed where MAV, VAR, AR and SE are chosen as the optimal feature set and linear discriminant analysis(LDA) is utilized to reduce the dimension. A combination of “pre-smoothed” and “post-smoothed” is applied to smooth the output in online recognition.The recognition of grasp force has been developed. Six levels of grasp force from 0 to 10 kilograms are designed to test the system. The performances of 10 features are analyzed to obtain the optimal feature set. Moreover, feature extraction, dimension reduction and curve fitting are combined to realize the online recognition of continuous grasp force.The overall system for the prosthetic hand has been build. LabVIEW is adopted to combine the myoelectric control system and the motor control system. Therefore, the prosthetic hand can perform prehensile postures and carry objects under the control of sEMG signals.