ファジィ・ニューラルネットワークを用いた上肢外骨格型パワーアシストロボットの食事動作認知アシストに関する研究

抄録

Recently, the aging problem is attracting great attention in Japan. Elderly or physically challenged people whose motor function has declined often have deteriorated environment cognitive abilities. Therefore, power-assist robots with perception-assist function have been proposed to assist such persons. In this paper, a fuzzy neural network method which makes the robot learns target position of the user for eating assist based on the user’s EMG signals, vertex position of hand motion, desk height and information of tool type is proposed. The effectiveness of the proposed method was evaluated by conducting experiments.