2A2-C10 自己組織化クラスタリングを用いた表面筋電位からの動作識別(リハビリテーションロボティクス・メカトロニクス)

概要

論文の詳細を見る
To discriminate motion using electromyographic (EMG) signals, generally we measure EMG signals and instruct user's desired motion to system directly using a keyboard, and training data is generated. But it's difficult to discriminate motion exactly with this method when the number of motion is increased, because the feature space of each motion's training data will be overlapped. Therefore, in this research, we proposed a method to classify EMG signals using self-organized clustering and create training data. Then, we conducted some experiments of motion discrimination using proposal method.
一般社団法人日本機械学会の論文
2007-05-11