Design and analysis of direct-action CMAC PID controller

This paper is to propose a direct-action (DA) cerebellar model articulation controller (CMAC) proportional-integral-derivative (PID) controller. The proposed controller, termed the DAC-PID controller, can generate four simple types of the nonlinear functions and then determine a control effort from those functions to control the process. In addition, the real-coded genetic algorithm is used to tune the parameters of the DAC-PID controller such that we can optimize those parameters. The performance of the proposed controller is also discussed in the sense of quantitative analysis. Simulation results demonstrate that the DAC-PID controller is superior to the conventional PID controller tuned by Ziegler-Nichols method and, moreover, as better as the optimal PID controller and the optimal fuzzy-PID controller.

[1]  Filson H. Glanz,et al.  Application of a General Learning Algorithm to the Control of Robotic Manipulators , 1987 .

[2]  C. Hang,et al.  Refinements of the Ziegler-Nichols tuning formula , 1991 .

[3]  Seppo J. Ovaska,et al.  A new CMAC neural network model with adaptive quantization input layer , 1996, Proceedings of Third International Conference on Signal Processing (ICSP'96).

[4]  James S. Albus,et al.  New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .

[5]  W. Thomas Miller,et al.  Real-time dynamic control of an industrial manipulator using a neural network-based learning controller , 1990, IEEE Trans. Robotics Autom..

[6]  George K. I. Mann,et al.  A systematic study of fuzzy PID controllers-function-based evaluation approach , 2001, IEEE Trans. Fuzzy Syst..

[7]  L.G. Bushnell,et al.  Report on the fuzzy versus conventional control debate , 1999, IEEE Control Systems.

[8]  S. Shahruz,et al.  Nonlinear PI Compensators That Achieve High Performance , 1997 .

[9]  R. Ortega.  Some remarks on adaptive neuro-fuzzy systems , 1996 .

[10]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[11]  George K. I. Mann,et al.  New methodology for analytical and optimal design of fuzzy PID controllers , 1999, IEEE Trans. Fuzzy Syst..

[12]  James S. Albus,et al.  Data Storage in the Cerebellar Model Articulation Controller (CMAC) , 1975 .

[13]  P. C. Parks,et al.  Convergence Properties of Associative Memory Storage for Learning Control Systems , 1990 .

[14]  George K. I. Mann,et al.  Analysis of direct action fuzzy PID controller structures , 1999, IEEE Trans. Syst. Man Cybern. Part B.

[15]  Chun-Shin Lin,et al.  Learning convergence of CMAC technique , 1997, IEEE Trans. Neural Networks.

[16]  Jing-Chung Shen,et al.  Fuzzy neural networks for tuning PID controller for plants with underdamped responses , 2001, IEEE Trans. Fuzzy Syst..

[17]  David E. Thompson,et al.  Neighborhood sequential and random training techniques for CMAC , 1995, IEEE Trans. Neural Networks.

[18]  A. Sideris,et al.  Learning convergence in the cerebellar model articulation controller , 1992, IEEE Trans. Neural Networks.

[19]  Cheng-Liang Chen,et al.  Design and analysis of a fuzzy logic controller , 1995 .

[20]  Fu-Chuang Chen,et al.  Practical stability issues in CMAC neural network control systems , 1996, IEEE Trans. Control. Syst. Technol..

[21]  Yau-Hwang Kuo,et al.  Hardware implementation of CMAC neural network with reduced storage requirement , 1997, IEEE Trans. Neural Networks.

[22]  Martin Brown,et al.  Comments on "Learning convergence in the cerebellar model articulation controller" , 1995, IEEE Trans. Neural Networks.