Basic research on an auxiliary system of Sit-to-Stand motion using lower limb EMG based on fast statisticl learning

2015 Fiscal Year Final Research Report

• Project/Area Number: 25350669
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Rehabilitation science/Welfare engineering
• Research Institution: The University of Tokushima
• Principal Investigator: Fukumi Minoru 徳島大学, ソシオテクノサイエンス研究部, 教授 (80199265)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 筋肉電位 / Simple-PCA / Simple-FLDA / 統計学習 / 足首動作

Outline of Final Research Achievements

In this research, a system to recognize ankle motions and to model a Sit-to-Stand motion using ankle EMG signals is developed for power-support devices. First, three motions of ankle, neutral, doesiflexion and plantar flexion using ankle EMG signals are recognized and 95% high accuracy is obtained for on-line experiments. Next, the Sit-to-Stand motion including three motions of ankle is recognized. The Sit-to-Stand motion is then divided into three motions. As a result, 62% accuracy is obtained and it is worse than that of three ankle motions. And also data distribution analysis in eigenspace showed that EMG data of Sit-to-Stand motion and that of three ankle motions measured at ankle are different. In the future, accuracy improvement including a normalization method for ankle EMG data for the Sit-to-Stand motion is needed.

Free Research Field

情報工学（ヒューマンセンシング，ソフトコンピューティング）