簡易檢索 / 詳目顯示

研究生: 陳宏謀
Hung-Mo Chen
論文名稱: 植基於小腦模型補償之狀態回授控制器設計
The Design of Full-State Feedback Controller based on Cerebellar Model Compensation
指導教授: 葉榮木
Yeh, Zong-Mu
學位類別: 碩士
Master
系所名稱: 工業教育學系
Department of Industrial Education
論文出版年: 2003
畢業學年度: 91
語文別: 中文
論文頁數: 167
中文關鍵詞: PID控制器小腦模型控制器狀態回授控制器
英文關鍵詞: PID Controller, Cerebellar Model Articulation Controller (CMAC), Full-state Feedback Controller
論文種類: 學術論文
相關次數: 點閱:159下載:74
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 受控體的數學模式,其建構存在著種種困難。本論文不是在探求精確的數學模式,而在尋求數學模型建構時所產生之誤差的補償方法。更何況傳統的控制方法對於一個複雜、較高階及非線性受控體而言,其控制器之設計及性能指標不易達成,縱然有精確的數學模式也徒然。所以本論文為了達成控制的目的,並考慮系統參數變動、負載干擾、受控體未考慮到的非線性因素以及數學模型建構時之誤差等等的影響,可能會導致系統的工作偏離其性能指標,因此對一複雜、較高階、非線性之受控體,提出一個<植基於小腦模型補償之狀態回授控制器設計>來達成控制的目的,並改善控制系統的性能指標。
    本研究採用狀態回授控制理論的系統設計,應用極點配置方法去穩定系統,並透過狀態回授來實現,使系統具有要求的閉迴路極點,再以外加小腦模型輔助補償控制的方式組成控制架構,因此而建立的控制方法所具備之特點為:對於二階之受控系統而言,由於狀態回授需要受控體的數學模式,吾人使用精度不太高的頻率響應法求取簡化模型,就能應用到控制系統中,而不必具有它們的數學描述;對於複雜、較高階及非線性之受控系統而言,雖然需要對系統的動態特性作數學描述,但是設計過程中無需對受控體或外部雜訊干擾等因素做精密測量及分析。而系統響應之誤差部份,則可藉由小腦模型控制器來改善。
    本研究,先使用Matlab語言撰寫模擬控制之程式,並比較PID控制器之傳統控制方法、PID控制器的改良型控制方法及本論文所提控制方法等之控制器性能,以性能指標和輸出響應圖來評估本論文所提之控制方法的控制品質;再將所設計之非線性受控體的控制方法,以Matlab程式模擬方式作倒單擺的倒立平衡控制及球軸系統的平衡控制,驗證本論文之控制架構能有效改善複雜、較高階及不穩定之非線性控制系統的控制性能(performance)。
    本研究為了驗證其控制方法具備學習與補償能力,於倒單擺的倒立平衡控制及球軸系統的平衡控制等之模擬當中,對負載干擾之抑制能力及控制系統參數變動之適應能力作測試,從小腦模型控制器加入與否的比較結果,驗證小腦模型控制器可改善控制系統並補償系統響應之誤差。於小腦模型控制器和可微小腦模型控制器的理論分析中,探討小腦模型控制器的特性。並以Turbo C語言撰寫等距離自動追蹤模型車控制系統程式,採取實驗方式以驗證小腦模型控制器之自動調整、自我學習的能力。

    There are various difficulties existing in the plant establishment with mathematics mode. This paper is not looking for the accurate mathematics mode but the error compensation method that is occurred during the process of finding the establishment of mathematics model. For a complicated, higher order and non-linear plant, the traditional control method will be limited in designing and hard to get its performance index. For achieving the controlling purpose, this paper takes the effect on parameter variation of the system, load interference, the non-linear factors that plant has not taken into consideration and the error that is resulted during the establishment of mathematics model into account and considers that the factors mentioned above may have caused the work of system to deviate from its performance index. Therefore, we design a controlling method for the complicated, higher order, non-linear plant and submit 「The Design of Full-state Feedback Controller based on Cerebellar Model Compensation」 to achieve the controlling purpose and to improve the performance index of the controller system.
    This study adopts the system design of the theory of full-state feedback controller and applies the pole-placement method to stabilize the system and to accomplish the stability through full-state feedback. The system will be possessed of the required pole of closed loop and then using the method of Cerebellar Model Assistance Compensation Feedback to form the control frame. Therefore, the established control frame has the following characteristics, which are: since full-state feedback needs the mathematics model of second order plant, we use the frequency response method with low accuracy to obtain the simplified model. And then it can be applied to the control system without their mathematics description. During the design procedure, we often use mathematics model to describe the dynamic properties of those complicated, higher order and non-linear plant.In the other word, we seldom take effort in accurately measuring and analyzing the plant and its external noise interference. And for the error in system response, it can be improved by using the Cerebellar Model Articulation Controller (CMAC).
    We use Matlab language to write the program of controlling method for simulation respectively and compare it with the traditional control method of PID controller, the improved model control of PID controller and the controller performance of control architecture mentioned in this study. And the performance index and output response diagram is used to estimate the control quality of control architecture that is mentioned in this study after being applied to the general plant system. To design the control method of non-linear plant, we use Matlab program to simulate the headstand balance control of the Inverted-Pendulum system and balance control of Ball-Beam system and prove that the control architecture of this study can be effectively applied to the complicated, higher order and instable non-linear control system.
    In the simulation of headstand balance control of the Inverted-Pendulum system and balance control of Ball-Beam system, the control ability of load interference and the adaptation ability of parameter variation of control system is tested and compared to the result with and without the addition of the CMAC. The compared result proves that the CMAC can effectively improve the control system and compensate the error of the system response. This paper discusses the characteristics of the CMAC through the discussion and theory analysis of one, two dimensions CMAC and differentiable CMAC. We use Turbo C language to write the control system program of equidistance auto tracing model car to prove the auto regulation and self-learning ability of the CMAC through experimental method.

    中文摘要 I 英文摘要 III 總目錄 VI 圖目錄 X 表目錄 XIV 第一章 緒論 1 1.1 研究背景與動機 1 1.2 研究目的 3 1.3 研究範圍與限制 3 1.4 研究方法 4 1.5 研究步驟 6 1.6 名詞釋義 9 第二章 文獻探討及理論基礎 10 2.1 控制系統的性能評價 10 2.2 小腦模型控制器 11 2.2.1 小腦模型控制器的基本架構 14 2.2.2 小腦模型控制器之記憶單元的分割方式 15 2.2.3 小腦模型控制器之回想與學習演算法 20 2.2.4 小腦模型控制器之學習流程 21 2.2.5 手算範例(以一維小腦模型控制器之學習為例) 23 2.2.6 小腦模型控制器學習範例與結果 26 2.2.7 可微分小腦模型控制器 30 2.2.8 CMAC與DCMAC之學習能力比較 42 2.2.9 小腦模型控制器與類神經網路的比較 45 2.3 極點配置設計狀態回授控制系統 46 2.3.1 連續時間系統的完全狀態可控性 46 2.3.2 狀態回授控制 48 2.3.3 極點配置 49 2.3.4 極點配置的設計步驟 52 2.3.5 極點配置的設計例 54 第三章 與其他控制方法相比較 56 3.1 求取控制對象之二階具時間延遲簡化模型 57 3.2 PID控制器的兩種設計方法與比較 60 3.2.1 Wang法設計PID控制器的參數 61 3.2.2 遺傳基因演算法求取PID控制器的參數 61 3.2.3 比較Wang法與遺傳基因演算法之控制品質 64 3.3 植基於小腦模型補償之狀態回授控制器控制 66 3.3.1 控制對象及其電腦Matlab程式處理方式 67 3.3.2 極點配置設計狀態回授控制 69 3.3.3 植基於小腦模型補償之狀態回授控制器的模擬輸出 71 3.4 本論文之控制方法與其他控制方法比較 75 第四章 應用於非線性受控系統 78 4.1 非線性受控系統的設計 78 4.2 倒單擺滑車系統 81 4.2.1 僅使用狀態回授控制器控制倒單擺滑車系統 87 4.2.2 植基於小腦模型補償之狀態回授控制器控制倒單擺 滑車系統 90 4.2.3 兩者應付環境變化之能力 94 4.3 球軸系統 97 4.3.1 僅使用狀態回授控制器控制球軸系統 101 4.3.2 植基於小腦模型補償之狀態回授控制器控制球軸 系統 104 4.3.3 兩者應付環境變化之能力 108 4.4 模擬結果之分析 111 第五章 等距離自動追蹤模型車控制系統實驗 113 5.1 系統規劃 113 5.2 硬體規劃 114 5.3 資料擷取卡的函數設定 119 5.4 感測距離的物理特性 121 5.5 控制系統的架構 123 5.6 控制系統實驗之軟體規劃 123 5.7 實驗結果 125 第六章 研究結論與建議 130 6.1 研究結論 130 6.2 研究建議 131 附錄一:求樣本狀態之超立方塊覆蓋編號(C++程式) 132 附錄二:求取控制對象之二階具時間延遲簡化模型 134 附錄三:PID控制器的架構 144 附錄四:Wang法求PID控制器的參數 146 附錄五:植基於遺傳基因演算法之PID控制器 148 附錄六:二階受控系統的極點配置設計,其電腦程式執行結果 154 附錄七:等距離自動追蹤模型車控制系統之電腦程式(Turbo C) 156 參考文獻 162 作者簡介 166

    [1] 沈金鐘,”PID控制器 理論、調整與實現”,滄海書局,90年8月初版.
    [2] J.G.Ziegler and N.B.Nichols,”Optimum settings for automatic controllers,”Trans.Amer.Soc.Mech.Eng.,Vol.64,pp.759-768,Novermber 1942.
    [3] 王威然(民89),”結合模糊與灰色控制理論實現小腦模型控制器之設計”,大同大學電機工程研究所碩士論文,pp1-3。
    [4] A.Thammano and C.H.Dagli, “A comparison of FAM and CMAC for nolinear control,”IEEE World Congress on Computational Intelligence.,Proceedings of the Third IEEE Conference, Vol.3,1994,pp. 1549-1553.
    [5] J. S. Albus, “A New Approach to Manipulator Control: The Cerebellar model articulation controller (CMAC),” J. Dynamic Syst., Meas., Contr., Trans. ASME, Series G, Vol.97, No.3, 1975, pp.220-227.
    [6] J. S. Albus, “Data Storage in the Cerebellar Model Articulation Controller (CMAC),” J. Dynamic Syst., Meas., Contr., Trans. ASME, Series G, Vol.97, No.3, 1975, pp.228-233.
    [7] ZHAO,Mingjie CHENG,Yiyu,”Fuzzy CMAC and Its Application in Function Learning”,Proceedings of the 3rd World Congress on Intelligent Control and Automation,June 28—July 2,2000,Hefei,P.R.China.
    [8] Dai Xiaolong, Zhou Xudong, A.Meio,T.Fukao,N.Adachi,”Optimal Design and Optimal Algorithm of Fuzzy CMAC”,SICE’98 July 29-31,China.,pp797.
    [9] M D Majors and R J Richards,”A Neural-Network-Based Flexible Assembly Controller”,’Artificial Neural Networks’,26-28 June 1995,Conference No.409, ©IEE,1995,pp268-273.
    [10] K. S. Hwang, C. S. Lin, “Smooth trajectory tracking of three-link robot: a self-organizing,” IEEE Trans. Systems, Man and Cybernetics, vol. 285, 1998, pp.680-692.
    [11] D.A.Handelman,S.H.Lane,“Integrating neural networks and knowledge-based systems for intelligent robotic control,” IEEE Control Systems Magazine, vol. 103, 1990, pp.77-87.
    [12] K. Y. Young, S. J. Shiah, “An approach to enlarge learning space coverage for robot learning control,” IEEE Trans. Fuzzy Systems, vol. 54, no. 4, 1997, pp.511-522.
    [13] J. J. Hu, G. Pratt, “Self-organizing CMAC Neural Networks and Adaptive Dynamic Control,” IEEE International Symposium on Control/Intelligent Systems and Semiotics 1999, Cambridge, MA., 1999, pp.259-265.
    [14] W. T. Miller, A. L. Kun, “Unified walking control for a biped robot using neural networks,” IEEE International Symposium on Intelligent Systems and Semiotics (ISAS) 1998, pp.283 -288.
    [15] A. L. Kun, W. T. Miller, “Adaptive dynamic balance of a biped robot using neural networks,” IEEE International Conference on Robotics and Automation 1996, vol. 1, pp.240-245.
    [16] Majors,M,Stori,J,Cho,D,”Neural Network Control of Automotive Fuel-Injection Systems”,1994,IEEE Cont.Sys.,14,31-36.
    [17] Junhong Nie and D.A.Linkens,”A Fuzzified CMAC Self-learning Controller”,0-7803-0614-7/93$03.00©1993IEEE, pp500.
    [18] W. Thomas Miller, Filson H. Glanz and L. Gordon Kraft, “CMAC: An Associative Neural Network Alternative to Back propagation,” Proceeding of the IEEE, Vol.78, No.10, 1990, pp.1561-1567.
    [19] W. T. Miller, “Real-Time Neural Network Control of A Biped Walking Robot,” IEEE Control Systems Magazine, Vol.141, 1994, pp.41-48.
    [20] Shelton R. O. & Peterson & J. K.,” Controlling a Truck with an Adaptive Critic CMAC Design. Simulation.”, 1992, 58, 5, 319-326.
    [21] J. S. Ker, Y. H. Kuo, R. C. Wen and B. D. Liu, “Hardware Implementation of CMAC Neural Network with Reduced Storage Requirement,” IEEE Trans. Neural Network, Vol.8, No.6, 1997, pp.1545-1556.
    [22] Y. Iiguni, “Hierarchical Image Coding via Cerebellar Model Arithmetic Computers,” IEEE Trans. Image Processing, 1996, Vol.5, No.10.
    [23] C. S. Lin, C. T. Chiang, “Learning Convergence of CMAC Technique,” IEEE Trans. Neural Networks, Vol.8, No.6, 1997, pp.1281-1292.
    [24] D. E. Thompson and S. Kwon, “Neighborhood Sequential and Random Training Techniques for CMAC,” IEEE Trans. Neural Networks, Vol.6, No.1, 1995, pp.196-202.
    [25] S. H. Lane, D. A. Handelman, J. J. Gelfand, “Theory and Development of Higher-Order CMAC Neural Networks,” IEEE Contr. Syst., vol. 12, 1992, pp. 23-30.
    [26] C. T. Chiang and C. S. Lin, “Integration of CMAC and Radial Basis Function Techniques,” IEEE International Conference on Intelligent Systems for the 21st, Vol. 4, 1995, pp.3263-3268.
    [27] C. T. Chiang and C. S. Lin, “CMAC with General Basis Functions,” Neural Network, vol.9, no.7, 1996, pp.1199-1211.
    [28] 周旭東、王國棟,模糊小腦模型神經網路,自動化學報,中國大陸1998,24(2),p173-177.
    [29] Chih-Ming Chen, Hahn-Ming Lee & Yu-Rong Hsieh (1999). “A New Learning Model of Hierarchical CMAC Neural Networks.”, Proceedings of Fourth National Conference on Artificial Intelligence and Applications, pp.17-22,1999.
    [30] Jar-Shone Ker, Yau-Hwang Kuo, Rong-Chang Wen & Bin-Da Liu,” Hardware Implementation of CMAC Neural Network with Reduced Storage Requirement.,” IEEE Transaction on Neural Networks, 8, 6, 1545-1556,1997.
    [31] 洪欽銘、陳志銘、羅維恆、黃昭諺。採用無失真壓縮技術精簡小腦模型控制器聯想記憶體之研究。第五屆人工智慧與應用研討會,pp.277-282,2000.
    [32] C. S. Lin, C. T. Chiang, “Learning Convergence of CMAC Technique,” IEEE Trans. Neural Networks, Vol.8, No.6, 1997, pp.1281-1292.
    [33] Katsuhiko Ogata, “Modern Control Engineering,”second edition,1990 by Prentice-Hall,Inc,pp.885-984
    [34] 須田信英,”PID 控制”,張道弘編譯,全華書局,1997年1月初版二 刷,pp.11-12.30-45
    [35] Ho W.K.,C.C.Hang and J.Zhou,(1997)”Self-tuning PID Control of a Plant with Under-damped Response with Specifications on Gain and Phase Margins,” IEEE Trans. Control Syst. Technol., Vol.5,pp.446-452
    [36] Shen J.C.,(2000)”New Tuning Method for PID Control of a Plant with Under-Damped Responses,”Asian Journal of Control,Vol.2,No.1,pp.31-41
    [37] Wang Q.G.,T.H.Lee,H.W.Fung,Q.Bi and Y.Zhang,(1999)”PID Tuning for Improved Performance,” IEEE Trans. Control Syst. Technol., Vol.7, No.4, pp457-465.
    [38] Mori Shozo ,〝Control of unstable mechanical system control of pendulum〞, International Journal Control , Vol.23 , No.5 , pp.673-692 , 1976.
    [39] Ren-Hou Li and Zhang Yi ,〝Fuzzy controller based on genetic algorithms〞, Fuzzy Sets and Systems , Vol.83 , pp.1-10 , 1996.
    [40] K. G. Eltohamy and C. Y. Kuo ,〝Real time stabilization of s triple link inverted pendulum using single control input〞, IEE Proceeding of Control Theory Application , Vol.144 , No.5 pp.498-504 , 1997.
    [41] 葉榮木,”自動控制-Visual Basic”,松崗電腦圖書公司,1998
    [42] John Hauser,Shankar Sastry,and Petar Kokotovic,”Nonlinear Control Via Approximate Input-Output Linearization:The ball and beam example,” IEEE Trans. On Automatic Control,Vol.37,No.3,pp.392-398.
    [43] 楊凱鈞(民90),”智慧型FUZZY控制實驗模組”,智控科技股份有限公司。
    [44] 楊慶祥(民91),”浮球位置控制系統之設計”, 國立台灣師範大學工業教育研究所碩士論文。

    QR CODE