基於EMG訊號之雙軸外骨骼機器人即時操控系統研發

Abstract

本論文提出利用 EMG 訊號即時操控雙軸外骨骼機器人。EMG 訊號，是肌 肉收縮時產生的生理訊號，經由分析系統可以推測使用者之意圖，而意圖分析本研究採用閥值分析法改良並搭配多種狀態的分析後，可解決 EMG 訊號不穩定的問題，進而達到即時操控外骨骼機器人，並且讓大多數使用者皆可輕易操作之目的。實驗結果顯示，所提出的 EMG 即時操控系統，確實能根據人的意圖隨時作動，且所有的受試者均能有效使用，對於醫療復健或者生活輔助的應用上都有所助益。

The object of this thesis is to use Electromyography (EMG) for real-time control of a Two-DOF upper-limb exoskeleton robot. EMG is a physiological signal generated during muscle contraction, and it can be used to derive user’s intention via analysis.The research tackles the uncertainties in EMG via an improved threshold approach. The proposed system can operate in real time, and be adapted to general user. From the experimental results, the Two-DOF upper-limb exoskeleton robot can be governed by user’s intention. It is potential applications lie on rehabilitation and daily activities.