EMG(electromyography)訊號是人體肌肉收縮而產生的一種類比生理訊號。此訊號會因為使用者的意圖而有不同的狀態，因此利用肌電訊號來作為控制義肢或機械輔具的控制命令，是一種直接有效的方法。針對肌電訊號與手臂動作之間關係和利用非侵入式偵測肌電訊號系統進行訊號分析。藉由單一表面電極擷取肌電訊號，進而判讀訊號的強度與複雜度，藉由變異數、標準差與近似熵，區分出目前人體動作意圖，以驗證可以利用最低限制控制機械輔具達到所要求之目的。另一個目的：針對整體系統來說，對同一肌群，當多個感測器其中之一故障，仍然可以藉由其他感測器執行原來系統的工作，而不使機械輔具因此延遲停宕。基於此原因，也意味著藉由完整分析訊號的特性，使得控制系統能夠更多元、多工化。 EMG (electromyography) signal is a physiological analog signal caused by muscle contraction. The signal is of a control order that will have the different states intended by the user and can be used to control the prosthesis or mechanical aids. It is a direct and effective method. Using relation between the EMG and the arm movement and non-invasion approach to detect the EMG signal, we can carry on the signal analysis. A single surface electrode is used to measure the EMG and the signal intensity and complexity are calculated by variance, standard deviation and approximate entropy to distinguish the current human motion intention which verifies the minimum limit control and can be used to achieve the required purposes of the mechanical aids. When one of sensors fails, the overall system for the same muscle groups can still work without letting the mechanical aids function abnormally. For this reason, it means that a complete analysis of the signal’s characteristics can have the control system be more multiple and versatile.
