Fuzzy Logic Controller For Leg Movement Classification Using sEMG Signal

Abstract

The Surface electromyography is the most favourable method to witness muscle activity. It involves no risk to
the subject as it is a non-invasive method. sEMG signals are processed and employed for rehabilitation engineering and
varied prosthetic technologies. In these days, sEMG signals are used for development of numerous controlling prototypes
based on gesture recognition modules. These modules distinguish different movements and utilise them to control a
machine. This work proposed a classification of knee extension at three levels using fuzzy logic technique. Surface
electromyography signals (sEMG) were acquired using hardware consisting of differential amplifier, non-inverting
amplifier, band pass filter and interface module from Vastus lateralis muscle which is responsible for leg extension
movements. MATLAB soft-scope was employed to import signals from hardware to system. For the task of classification,
fuzzy logic controller was used. For signal analysis three parameters, Root Mean Square, Median and Standard
Deviation were selected as inputs to fuzzy logic controller. Results showed that out of all three parameters, standard
deviation was proved to be the best parameter for discriminating movements.