Toward the identification of EMG-signal and its bio-feedback application

Abstract

The electromyogram (EMG) is known to be induced by certain muscle contraction. Except for those disabled who totally lost their extremity, surface EMG signals can be measured from the skin surface. Due to this property, EMG can detect the muscle contraction, and thus can be applied to estimate the required torque for orthosis, and it is thus possible to operate the orthosis in real-time by bio-feedback to enhance human's physical ability instead of using an artificial limb. In this article, firstly, we introduce inverse dynamics and a muscle mechanics model to infer EMG which is related to the knee torque. Secondly, we make use of an adaptive neuro-fuzzy inference system (ANFIS) to establish a model that related EMG and knee torque. Lastly, we will devise an EMG-based bio-feedback lower limb othosis in real-time through several experiments to validate its efficacy.