A sEMG-based Real-time Adaptive Joint angle Estimation and Control for a Prosthetic Hand Prototype

摘要

This paper presents a novel approach of a surface electromyographic (sEMG)-based, real-time Model Reference Adaptive Control (MRAC) strategy for joint angle estimation and control of a prosthetic hand prototype. The proposed design is capable of decoding the pre-recorded sEMG signal as well as the sensory force feedback from the sensors to control the joint angle of the prosthetic hand prototype using a PIC 32MX360F512L microcontroller. The input sEMG signal is rectified, filtered using a Half-Gaussian filter and fed to a Kullback Information Criterion (KIC) based fusion algorithm, which give the overall estimated joint angle (position). This estimated angle is given as input to the prosthetic hand prototype. Then the MRAC along with a two stage proposed embedded design is used for the position control of the prosthetic hand. For the realtime operation, the measured angle data at the joint is fed back to the controller. The data is transmitted to the computer through a universal asynchronous receiver/transmitter (UART) interface of the proposed embedded design. The experimental results in real time show good performance in controlling the angle of the prosthetic hand prototype.