摘 要
功能性神經電刺激術主要用於恢復中樞神經與周邊神經間失去聯繫之病患，為神經工程主要方法。此法利用植入式電極如銬型電極與周邊神經接觸，由電刺激器控制電流強度與脈寬使癱瘓之肌肉產生收縮力，但由於受刺激之肌肉之徵召曲線會隨時間而變化且肌肉容易疲勞。本研究之目的即在於發展一套能同時控制電流強度與脈寬之方法，同時配合PID和三種模糊控制器以進行家兔踝關節扭矩控制。將家兔踝關節在自然角度及不同之角度下進行等長扭矩電刺激的結果顯示：（1）Fuzzy PD響應快，超越量小但干擾靈敏度高且穩態誤差太高（2）Fuzzy PD+I響應快超越量小且穩態誤差小（3）Fuzzy PI穩態誤差最小，控制訊號平滑，安定時間短，但超越量稍大。三者中以Fuzzy PI之ITAE值最小。總之，本研究發展之簡化電流脈寬調變法可以避免受刺激肌肉快速疲勞，Fuzzy PI控制較PID控制優且適用於踝關節在運動範圍內之等長力矩控制。

Abstract
The functional neuromuscular stimulation (FNS) is a major method in neural engineering for restoring the patients who lose the connection between central and peripheral nervous systems. IN FNS an implanted electrode, e.g. cuff electrode contacts with the peripheral nerve and the amplitude and pulse-width of the stimulus current is controlled by an electrical stimulator. Due to the time-varying characteristics of muscle recruitment and fast fatigue of artificially stimulated muscles, it is necessary to develop a method which can simultaneously control the amplitude and pulse-width of the stimulus and combined with feedback control algorithms to yield better FNS. The goal of this thesis is to develop a simplified amplitude/pulse-width modulation (SAWM) technique and combined with a PID and three fuzzy controls, namely, Fuzzy PD, Fuzzy PI, Fuzzy PD+I to the isometric ankle torque control of anesthetized rabbits. The rabbit’s ankle was fixed at natural and various positions in the range of motion and different control algorithms were employed to test the step response of the FNS system. Experimental results showed that: (1) Fuzzy PD has fast response, small overshoot percentage but high sensitivity to disturbance and large steady-state error (2) Fuzzy PD+I has fast response, small overshoot percentage and small steady-state error (3) Fuzzy PI has smallest steady-state error, smaller settling time, smoother control signal but large overshoot percentage. For the integrated time multiplied by absolute error (ITAE), the Fuzzy PI is superior to others. From the results, one may conclude that the SAWM technique is effective in preventing the fatigue of stimulated muscles. Fuzzy PI control is better than PID control and it is adaptive to the ankle position for isometric torque control on anesthetized rabbits.

參考文獻
[1] W. L. C. Rutten, H. J. Wier, and J. H. M. Put, “Sensitivity and selectivity of intraneural using a silicon electrode array,” IEEE Trans. Biomed. Eng., vol. 38, no. 2, pp. 192-196, 1991.

[2] M. S. Ju, H. C. Chien, G. S. Chen, C. C. K. Lin, C. H. Chang, and C. W. Chang, “Design and fabrication of multi-microelectrode array for neural prosthesis,” Chin. J. Med. Biol. Eng., vol. 22, pp. 33-40, 2002.

[3] A. Q. Choi, J. K. Cavanaugh, and D. M. Durand, “Selectivity of multiple-contact nerve cuff electrodes: a simulation analysis,” IEEE Trans. Biomed. Eng., vol. 48, pp. 165-172, 2001.

[4] D. J. Tyler and D. M. Durand, “Functionally selective peripheral nerve stimulation with a flat interface nerve electrode,” IEEE Trans. Neural. Syst. Rehabil. Eng., vol. 10, pp. 294-303, 2002.

[5] A. Kralj and T. Bajd, “Functional electrical stimulation: standing and walking after spinal cord injury” Boca Raton, FL: CRC, 1989.

[6] R. Davoodi and B. J. Andrews, “FES standing up in paraplegia: a comparative study of fixed parameter controllers,” IEEE in Med. Biol. Society, vol. 1, pp. 447-448, 1996.

[7] J. J. Abbas, “Feedback control of coronal plane hip angle in paraplegic subjects using neuromuscular stimulation,” IEEE Trans. Biomed. Eng. vol. 38, pp. 678-698, 1991.

[8] N. Lan, P.E. Crago, and H. J. Chizeck, “Control of the end-point force of a multijoint limb by functional electrical stimulation,” IEEE Trans. Biomed. Eng. vol. 38, pp. 953-965, 1991.

[9] N. Itakura, K. Fujita, K. Kubo, Y. Iguchi, and H. Minamitani, “Evaluation of FES control system employing adaptive and PI controllers,” IEEE in Med. Biol. Society, vol. 4, pp. 1738-1740, 1988.

[10] R. Davoodi, and B. J. Andrews, “Computer simulation of FES standing up in paraplegia: a self-adaptive fuzzy controller with reinforcement learning,” IEEE Trans. Rehabil. Eng. vol. 6, pp. 151-161,1998.

[11] M. S. Hatwell, B. J. Oderkerk, C. A. Sacher, and G. F. Inbar, “The development of a model reference adaptive controller to control the knee joint of paraplegics,” IEEE Trans. Automat. Contr. vol. 36, pp. 683-691, 1991.

[12] P. Bonato, S. H. Roy, M. Knaflitz, and C. J. De luca, “Time-frequency parameters of the surface myoelectric signal for assessing muscle fatigue during cyclic dynamic contractions,” IEEE Trans. Biomed. Eng., vol. 48, pp. 745-753, 2001.

[13] P. Mela, P. H. Veltink, P. A. Huijing, S. Salmons, J. C. Jarvis, “The optimal stimulation pattern for skeletal muscle is dependent on muscle length,” IEEE Trans. Neural. Syst. Rehabil. Eng., vol. 10, pp. 85-93, 2002.

[14] H. Qi, D. J. Tyler, and D. M. Durand, “Neurofuzzy adaptive controlling of selective stimulation for fes: a case study,” IEEE Trans. Rehab. Eng., vol. 7, pp. 183-192, 1999.

[15] H. S. Cheng, M. S. Ju, and C. C. K. Lin, “Estimation of peroneal and tibial afferents from a multi-channel cuff placed on sciatic nerve,” Muscle & Nerve, vol. 32, pp. 589-599, 2005.

[16] C. C. Lee, “Fuzzy logic in control system: Fuzzy logic controller, Part I and Part II,” IEEE Trans. System, Man and Cyb., vol. 20, pp. 404-435, 1990.

[17] J. Jan, “Tuning of fuzzy PID controllers,” report no 98-H 871(fpid), 1998. http://www.iau.dtu.dk/~jj/pubs/fpid.pdf

[18] H. J. Chizeck, P. E. Crago, and L. S. Kofman, “Robust closed-loop control of isometric muscle force using pulsewidth modulation,” IEEE Trans. Biomed. Eng. vol. 35, pp. 510-517, 1988