前腕部電動義手を制御するための筋電位解析システムの開発

抄録

We have studied about an EMG analysis system for an electric prosthetic forearm control. Our previous experiment for motion recognition yielded high recognition accuracy results. This experiment with the 48-channel EMG system used Monte Carlo Method to decide sensing electrode position, and linear discriminant analysis. And, we have found when four channel electrodes are attached to correct positions on the forearm, it is able to recognize motions of it. This objective of this study is the development of a wearable the four-channel EMG analysis system by using our previous experimental results. Therefore, we have done motion recognition experiment, to confirm that obtain high recognition accuracy even if 4-channel EMG system sensing at electrode positions determined by 48-channel EMG system. In conclusion, the present study has demonstrated that high recognition accuracy requires training for using EMG system, and we found out sufficient training period depend on examinee.