Voice and EMG signal based low-cost automated assistive arm

Abstract

Utilization of electromyogram (EMG) signal is one of the most utilitarian and practical methods of developing a prosthetic or orthotic limb. In addition to EMG, voice recognition method can also be convenient and applicable in specific situations. In this article, a pragmatic and cost effective method of constructing a prosthetic arm using the mentioned two different signals is presented. Post-filtering was performed on the EMG signal and necessary features were extracted and evaluated while the voice input was processed by the specific Android application. The developed prosthetic arm has been able to perform standard tasks such as extension and grabbing things that are out of reach with appreciable success. Experiments were carried out to evaluate the performance and user-friendliness of the developed arm and the results suggest that this prosthetic arm has the potential to assist in the day-to-day activities of persons with disabilities or below-elbow amputees.