A neuro-fuzzy control system based on feature extraction of surface electromyogram signal for solar-powered wheelchair

This paper presents the design and implementation of a low-cost solar-powered wheelchair for physically challenged people. The signals necessary to maneuver the wheelchair are acquired from different muscles of the hand using surface electromyography (sEMG) technique. The raw sEMG signals are collected from the upper limb muscles which are then processed, characterized, and classified to extract necessary features for the generation of control signals to be used for the automated movement of the wheelchair. An artificial neural network-based classifier is constructed to classify the patterns and features extracted from the raw sEMG signals. The classification accuracy of the extracted parameters from the sEMG signals is found to be relatively high in comparison with the existing methods. The extracted parameters used to generate control signals that are then fed into a microcomputer-based control system (MiCS). A solar-powered wheelchair prototype is developed, and the above MiCS is introduced to control its maneuver using the sEMG signals. The prototype is then thoroughly tested with sEMG signals from patients of different age groups. Also, the life cycle cost analysis of the proposed wheelchair revealed that it is financially feasible and cost-effective