A Neuro-Fuzzy System for Classifying Fatigue Degree of Wheelchair User

With the increase of disabled people, the functionalities of smart wheelchair as a mobility-assisted equipment are being more and more enriched and extended. However, fatigue detection for wheelchair users is still not explored widely. This paper proposes a complete system and approach to classify fatigue degree for manual wheelchair users. In our system, physiological and kinetic data are collected in terms of sEMG, ECG, and acceleration signals. The necessary features are then extracted from the signals and integrated with self-rating method to train a neuro-fuzzy classifier. Finally, four degrees of this fatigue status can be distinguished by our system; this can provide further fatigue prediction and alertness in case of musculoskeletal disorders (MSD) caused by underlying fatigue.