Enhanced impedance control for an sEMG-based prosthetic using RNE approach

Abstract

Impedance control has been a major challenge faced by most robotic limbs and arms. One of the ways of enhancing the prosthetic performance is by optimising the kinematic and dynamic systems. Therefore, based on surface electromyography (sEMG) signal, this study implemented the recursive Newton-Euler (RNE) approach to establish an effective control for a prosthetics. The model introduced a primal approach which trains the subject to utilise the prosthetic sEMG data through the use of an operator-machine communal control. Further, the study utilised the estimated kinematics of the prosthetic manipulator to measure the joint torques which is expected to activate the prosthetic robotic arm, while using the active compensation to mask the real dynamics. Finally, the research used several simulations for prosthetic grasping, control and angular velocities to show that with the incorporation of the RNE approach, impedance is efficiently managed while the prosthetic devices are better controlled by the subjects.