2017 年 83 巻 851 号 p. 17-00175
In this paper, we propose a system to identify the intention of motion based on the change of the surface shape of the upper arm. The proposed system uses six photoreflectors to measure the change of the surface shape of the forearm close to elbow joint. The changes of the surface shape of the forearm close to elbow joint are caused by muscle contraction. For this reason, the proposed system aims to be mounted on prosthetic hands for upper limb amputees who have residual muscular. Conventional identification methods of the intention of motion using photoreflectors were performed with a single motion such as grasping / releasing. In contrast, the proposed system identifies the six types of composite motions of hand and forearm. Complex processing is required to identify the six types of composite motions at a time. In the proposed system, it is possible to identify the composite motion by relatively simple multiple regression model, by separating into the motion of the forearm and the hand. In order to confirm the effectiveness of the proposed system, we conducted a discrimination experiment of composite motion by six volunteers and a discrimination experiment of composite motion at long-time use by one volunteer. As a result of the experiment by six volunteers, the average discrimination rate was 94.9 %. Moreover, the average discrimination rate of long-time use was 97.8 %. From these experimental results, we showed the possibility of discrimination of composite motion using photoreflectors.