A Virtual Training Environment for Prosthetic Control

Abstract

One factor that impacts the operation and functionality of prosthetic devices is the lack of sensory feedback. Currently, there are no commercial prosthetics that offer feedback. To study the effects of feedback, a real-time a system, based on the Matlab toolbox SynGrasp, is implemented in Simulink (The Mathworks, Inc.). SynGrasp supports the simulation of a variety of human and robotic hands. For the purposes of this study, a 20 degrees of freedom hand was selected as it accurately represents all hand movements associated with a circular grasp. The toolbox can simulate hand-object grasp interaction and calculate the contact forces. This functionality is ported into Simulink for real-time use. EMG control is achieved using two active bipolar SEMG electrodes on the ventral and dorsal surfaces of the forearm to control finger flexion and extension respectively. Sensory feedback is incorporated as a vibration at 235 Hz, via an 8 mm linear resonant actuator (Precision Microdrives Ltd.), secured over the lateral side of the upper arm. The mean contact force magnitude from the virtual environment is mapped to the vibration amplitude generated by the vibrotactile device. It has been shown that feedback in prosthesis control improved grasp precision and speed of grasping task completion. As well, it has been shown that a complex control scheme has a negative impact on subject performance. The goal of this study is to develop and test a relatively simple virtual tool for grasp training, with and without vibrotactile feedback.