EMG信号とロボットハンド操作のダイレクトマッピング

抄録

The objective of this paper is to show an efficient method of representing the continuous transition from the flexed fingers of a fist to the stretched state of an open palm and vice versa by using the surface electromyographic (hereafter EMG) data provided by a device worn on the forearm. Using the difference between the EMG linear envelopes of the muscles 'Musculus extensor carpi ulnaris' and 'Musculus flexor digitorum profundus' it is possible to represent the current state of the hand, from open to closed, with a single, analogue signal. In this paper, we show that a simulated robotic hand can be controlled in real-time by processing human myoelectric signals based on our method. To show a validity of our method, we present experimental results.