[8a-N304-1~11] FS.1 Focused Session "AI Electronics"

11:00 AM - 11:15 AM JST(2:00 AM - 2:15 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[8a-N304-8] Improvement of Motion Readout Capacity in Myoelectric Signal Analysis Inspired by Reservoir Computing

〇Yusuke Hoshika¹, Zenji Yatabe¹, Seiya Kasai¹ (1.RCIQE, Hokkaido Univ.)

Keywords:

Myoelectric prosthesis hand,Surface myoelectric signal analysis,Reservoir computing

To improve controllability of myoelectric prostheses, we have developed a unique myoelectric signal analysis technique considering the similarity between reservoir computing (RC) and the human motor control system. An issue in our technique is that the system could not respond to a hand motion, grasp. In this paper, we figured out and solved the issue. We found that myoelectric signal involved in grasp motion was attenuated because the muscle involved in the grasp is deep inside of the forearm. To solve this problem, we applied logarithmic transformation to selectively amplify the myoelectric signal. As a result, the system successfully became to respond grasp motion.