Development of a functional electrical stimulation feedback control system using infrared muscle activity sensor

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K17795
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 61020:Human interface and interaction-related
• Research Institution: Kochi National College of Technology
• Principal Investigator: Yoshioka Masataka 高知工業高等専門学校, ソーシャルデザイン工学科, 准教授 (80805804)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 赤外光 / 指先動作推定 / 電気刺激 / 機械学習 / LSTM

Outline of Final Research Achievements

This study aims to develop a muscle feedback control system using Functional Electrical Stimulation (FES) and Infrared Muscle Activity Sensors (IrMAS). It involves estimating fingertip movements through machine learning based on IrMAS and stimulating antagonistic muscles with multiple FES applications. The results show a correlation between muscle brightness obtained from echo motion images and IrMAS, proving that IrMAS is related to muscle movement. Machine learning with RNN and LSTM achieved a maximum correlation coefficient of r = 0.93 for estimating fingertip movements. Additionally, a relationship between stimulation voltage and frequency and fingertip exertion force was observed in FES, and multiple stimulations achieved antagonistic muscle movements. Moving forward, these findings will be advanced towards real-time implementation to realize feedback control.

Free Research Field

生体医工学

Academic Significance and Societal Importance of the Research Achievements

機能的電気刺激による筋肉の制御は，筋電位を入力としてパワーアシストを行う装置は存在するが，あくまで筋収縮の補助であり，直接的なフィードバック制御は実現できていなかった。この問題に対し，赤外光による安全で安価な測定手法のIrMASを導入することにより，
電気的ノイズの問題が解決され，使用者に合わせた刺激が可能となる。また，IrMASにおける指先動作推定では，非間接的に指の動きを推定する既存のカメラによる指先推定と比べ，手を画角に収める必要もなく，筋電位と比べても弱い力も検出出来るため，様々なインターフェースへの応用も期待できる。