Evaluation of Fuzzy-Neuro Modifiers for Compensation of the Effects of Muscle Fatigue on EMG-Based Control to be Used in Upper-Limb Power-Assist Exoskeletons

Abstract

Assistive robotic technology can play a major role to improve the quality of life of the physically weak people such as aged, injured, disabled or handicapped. Many assistive devices have been developed according to the needs of such individuals. Especially, upper-limb power-assist exoskeletons have been able to draw attention, as the upper limb motions are very important for the daily activities. Electromyography (EMG) signals of the upper limb muscles have frequently been used as a primary signal to control the upper-limb power assist exoskeletons, because the EMG signals directly reflect the motion intention of the user. However, one of the main problems for EMG-based control is the muscle fatigue, because the muscle fatigue can affect the EMG patterns. When the user's muscles get fatigued, it is required to consider the variety of EMG signals on the EMG-based controllers. In this paper, the effects of muscle fatigue on EMG-based control are analyzed based on upper-limb elbow flexion/extension motions and fuzzy-neuro modifiers which intend to compensate for the muscle fatigue effects based on a combination of EMG Root Mean Square (RMS) and EMG Mean Power Frequency (MPF) are proposed. The effectiveness of the proposed fuzzy-neuro modifiers for compensation of the effects of muscle fatigue are evaluated by conducting experiments. Key