Hardware Implementation of CMAC Neural Network using FPGA Approach

This paper proposes a hardware design and implementation of a cerebellar model articulation controller (CMAC neural network). The computer software simulation and FPGA hardware realization of CMAC neural network are developed. The Altera series FPGA chips are used to implement CMAC to achieve the characteristics of small size, fast execution speed and less memory. The system accuracy is verified through software simulation and hardware test. Two nonlinear functions and a real-time photovoltaic system are employed to illustrate the system performance. From the experimental results, it can be verified that the designed FPGA-based CMAC neural network can accurately model the nonlinear systems.

[1]  Dong-Il Cho,et al.  CMAC neural network controller for fuel-injection systems , 1995, IEEE Trans. Control. Syst. Technol..

[2]  Chih-Min Lin,et al.  Neural-network hybrid control for antilock braking systems , 2003, IEEE Trans. Neural Networks.

[3]  Kourosh Danai,et al.  Model-based recurrent neural network for modeling nonlinear dynamic systems , 1999, Proceedings of the 1999 IEEE International Conference on Control Applications (Cat. No.99CH36328).

[4]  G. Horvath,et al.  FPGA implementation of the kernel CMAC , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[5]  Chih-Min Lin,et al.  Adaptive CMAC-based supervisory control for uncertain nonlinear systems , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[6]  Shun-Feng Su,et al.  Credit assigned CMAC and its application to online learning robust controllers , 2003, IEEE Trans. Syst. Man Cybern. Part B.

[7]  Rong-Jong Wai,et al.  Adaptive hybrid control for linear piezoelectric ceramic motor drive using diagonal recurrent CMAC network , 2004, IEEE Transactions on Neural Networks.

[8]  Tung-Sheng Chiang,et al.  A simple and converged structure of addressing technique for CMAC/spl I.bar/GBF , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[9]  Frank L. Lewis,et al.  Multilayer neural-net robot controller with guaranteed tracking performance , 1996, IEEE Trans. Neural Networks.

[10]  Lin Chun-Shin,et al.  CMAC with General Basis Functions. , 1996, Neural networks : the official journal of the International Neural Network Society.

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