EMG prosthetic hand controller discriminating ten motions usingreal-time learning method

摘要

We discuss the necessity of a learning mechanism for an EMGprosthetic hand controller, and the real-time learning method isproposed and designed. This method divides the controller into threeunits. The analysis unit extracts useful informations for discriminatingmotions from the EMG. The adaptation unit learns the relation betweenEMG and control command and adapts operator's characteristics. Thetrainer unit makes the adaptation unit learn in real-time. Experimentsshow that the proposed controller discriminates ten forearm motions,which contain four wrist motions and six hand motions, and learns within4~25 minutes. The average of the discriminating rate is 91.5%