Abstract
Myoelectric hand is a prosthetic hand to perform the estimated motion based on the myoelectric potential generated from the human arm, therefore the precise motion recognition is essential. Then, myoelectric potential tends to change in arm posture change. This study proposes a motion recognition method using fuzzy reasoning in response to changes in the myoelectric potential due to the arm postures change. The polynomial regression is applied to the measured myoelectric potential in order to represent the posture change characteristics, and fuzzy reasoning with the membership function based on the polynomial approximation and standard deviation is utilized for the motion recognition. Some experiment results are provided to show the effectiveness of the proposed method.
