Abstract
Power-assist of the lower-limb movement is sometimes important for the physically weak persons such as elderly, disabled, or injured persons. A power-assist exoskeleton robot which is controlled based on user's electromyogram has been developed to realize the assist according to the user's movement intention. In this paper, we propose an EMG-based control method which provides the required joint torque for lower-limb power-assist to realize the posture of deep knee flexion in Japanese lifestyle. The proposed controller can be automatically trained with fuzzy-neurro modifier.