肌電訊號是一種生理現象，伴隨著人體肌肉的收縮會以成正比的趨式呈現在人體皮膚表面。運用此種現象，我們可以用來設計輔具的輸出。除非是肢體完全殘缺者，只要有相對於不出力時能呈現較大的肌電訊號，我們就可以藉此告訴系統該肌肉已經有作動的意圖，而輔具可以開始輸出輔助力矩於特定部位。因而此方向的輔具設計屬於「強化使用者體能型」，而非為肢體殘缺者設計的「義肢型」。
本文藉由探討肌電訊號的本質，重新設計一個前置處理電路，包含放大與濾波的功能，讓擷取的訊號在解析度與辨識度上都符合使用，接著進一步以實驗建立起大腿股四頭肌的肌電訊號，與人體執行從坐姿到站姿動作的相關性，找出肌電訊號與膝蓋所需支撐力矩的關係，設計一套即時運算方法，將接收到的肌電訊號轉化成輔具的輸出力，並施加在膝蓋上，以期讓使用者在任何膝蓋彎曲角度下都能夠如直立般輕鬆。
本實驗乃國科會計畫的一部分，扮演的角色為交接與改良的部分。前一代的輔具輸出依據是「測力鞋-角度計模組」的運算結果，這在本文中會經常提及，並將以「原始方法」代稱。本文所提之肌電訊號回饋，既定位為改良的角色，勢必要將使用本方法之前與之後作比較。故介紹先前的方法與為何本文提出的方法可以有效改良「原始方法」的表現，在文中也會經常被提及，最後則以實驗來驗證。

Abstract
Electromyogram (EMG) is an phenomenon of physiology. As the
muscle contracting more, the EMG signal grows shaper. We can use this
kind of phenomenon designing our prosthesis. Except for those injured
who totally lost their entire extremity, we can gather data from their EMG.
The prosthesis system will notice that the user is tend to move his(her)
body when the shacking amplitude of the use's EMG is much greater than
that in the static release, static condition. Therefore, the object of
designing this prosthesis is to enhance human's physical ability instead of
providing artificial limbs.
In the beginning of the thesis, we introduce the EMG processing
circuit, which contain amplifiers and filters. The processed EMG data
will then be connected to the torque on knee during the sit- to- stand
motion. While the EMG signal is received by system during the time the
user bend his knee, the relative assist torque will come out immediately,
trying to make the user feel as easy as just stand still.
The objective of this study is trying to improve the original assistant
force algorithm, which is based on the force-estimate shoe-and-angle
meter module. This kind of method will be mentioned usually in this
thesis and kept preparing with the advanced method proposed in this
thesis.

[1] M.Asghari Oskoei and H.Hu ” Myoelectric control systems—A survey,” Biomedical Signal Processing and Control 2, pp 275–294, 2007.
[2] G.-W. Choi, I.-H. Moon, G.-H. Choe and M.-S. Mun,” Development of Surface Myoelectric Sensor for Myoelectric Hand Prosthesis,” Power Electronics Specialists Conference, 2006. PESC '06. 37th IEEE, pp 1-5 , 18-22 June 2006,
[3] R.S. Ochi and P.R. Cavanagh, “Reliability of surface EMG measurements over 12 hours,” ELSEVIER, Journal of Electromyography and Kinesiology ,17, pp 365–371 ,2007.
[4] T. W. CALVERT, and A. E. CHAPMAN,” The Relationship Between the Surface EMG and Force Transients in Muscle: Simulation and Experimental Studies,” Journal of Electromyography and Kinesiology 17 , pp 682-689 ,1977.
[5] J.-U. Chu, I. Moon, S.-K. Kim,and M.-S. Mun, ”Control of Multifunction Myoelectric Hand Using a Real-Time EMG Pattern Recognition,” Intelligent Robots and Systems, 2005. (IROS 2005). 2005 IEEE/RSJ International Conference , pp 3511- 3516, 2005.
[6] T. F. Jiangang, L. Z. WANG and Y. SUN, “An Image Compressing Algorithm based on PCA/SOFM Hybrid Neural Network,” Industrial Electronics Society, 2003. IECON '03. The 29th Annual Conference of the IEEE, pp 2103-2107, 2003.
[7] G. M. Foody, “Applications of the self-organising feature map neural network in community data analysis,” Ecological Modelling ,120, pp 97–107, 1999.
[8] S. Lin J. Si,” Convergence Properties of SOFM Algorithm for Vector Quantization,” 1997 IEEE International Symposium on Circuits and Systems, June 9-12, Hong Kong, pp 509-512, 1997.
[9] Y. Kun, Z. Hong and Y.-J. Pan,” Human Face Detection Based on SOFM Neural Network,” Information Acquisition, 2006 IEEE International Conference on, pp 1253-1257, 2006.
[10] J.-U. Chu, , I. Moon, Y.-J. Lee, S.-K. Kim, and M.-S. Mun” A Supervised Feature-Projection-Based Real-Time EMG Pattern Recognition for Multifunction Myoelectric Hand Control,” IEEE/ASME transactions on mechanics, vol. 12, no.3, pp 282-290, 2007.
[11] A. O. Andradea , S. J. Nasutob and P. Kyberd,” Extraction of motor unit action potentials from electromyographic signals through generative topographic mapping,” Journal of the Franklin Institute, 344, pp 154–179, 2007
[12] E. J. Scheme, B. Hudgins, and P. A. Parker, “Myoelectric Signal Classification for Phoneme-Based Speech Recognition,” IEEE transactions on biomedical engineering, vol.54, no.4, pp 694-699,2007
[13] A.-C. Dupont and E. L. Morin, “A Myoelectric Control Evaluation and Trainer System,” IEEE transactions on rehabilitation engineering, vol.2, no.2, pp 100-107, 1994
[14] K. Terashima , K. Mouri, P. Minyong ,H. Kitagawa and T. Miyoshi “Identification and hybrid impedance control of human skin muscle by multi-fingered robot hand,” Intelligent Robots and Systems, IROS 2007. IEEE/RSJ International Conference, pp 2895-2900, 2007
[15] J. Romkes and R. Brunner “An electromyographic analysis of obligatory(hemiplegic cerebral palsy) and voluntary (normal)unilateral toe-walking” ELSEVIER,Giat & Posture, 26 , pp 577–586 , 2007.
[16] http://www.cyberdyne.jp/english/robotsuithal/move.html
[17] S.Lee and Y.Sankai,” Power assist control for walking aid with HAL-3 base on EMG and impedance adjustment around knee joint,” Intelligent Robots and System, 2002. IEEE/RSJ International Conference, vol. 2, pp.1499 - 1504, 30 Sept.-5 Oct. 2002.
[18] Y. Hasegawa, Y. Mikami, K. Watanabe and Y. Sankai,” Five-Fingered Assistive Hand with Mechanical Compliance of Human Finger,” IEEE International Conference on Robotics and Automation, pp 718-724, 2008.
[19] S. N˜omm, E. Petlenkov , J. Vain , J. Belikov , F. Miyawaki and K. Yoshimitsu, ” Recognition of the Surgeon’s Motions During Endoscopic Operation by Statistics Based Algorithm and Neural Networks Based ANARX Models,” Proc. of 17th IFAC World Congress (Seoul, July 2008), pp 14773-14778, 2008.
[20] T. Nakai, S. Lee,H.Kawamoko and Y.Sankai ” Development of power assistive leg for walking aid using EMG and Linux,” Asia symposium industrial automation and robotics (ASIAR2000) , pp 1295-1310, 2000.
[21] C. Lee, The relationship between isometric EMG and torque of the quadriceps femoris before and after fatigue, Graduated School of Physical Education Doctoral Dissertation , NTNU, 1998