Title :
Radial basis function neural network for regulation of nonlinear systems
Author :
Kostanic, Ivica N. ; Ham, Fredric M.
Author_Institution :
Florida Inst. of Technol., Melbourne, FL, USA
Abstract :
A large class of nonlinear discrete systems with accessible states can be controlled through feedback linearization. This paper develops a practical algorithm for state feedback control design using radial basis function neural networks (RBFNN). Linear least-squares is coupled with a Gram-Schmidt orthogonalization procedure to perform size reduction of the neural networks. An example of regulating a nonlinear plant is included to illustrate the effectiveness of the proposed algorithm
Keywords :
discrete systems; feedforward neural nets; least squares approximations; multilayer perceptrons; nonlinear control systems; state feedback; Gram-Schmidt orthogonalization; accessible states; feedback linearization; linear least-squares; nonlinear discrete systems; nonlinear plant; nonlinear systems regulation; practical algorithm; radial basis function neural network; size reduction; state feedback control design; Control systems; Counting circuits; Neural networks; Neurofeedback; Nonlinear equations; Nonlinear systems; Radial basis function networks; Regulators; State feedback; Vectors;
Conference_Titel :
Southeastcon '96. Bringing Together Education, Science and Technology., Proceedings of the IEEE
Conference_Location :
Tampa, FL
Print_ISBN :
0-7803-3088-9
DOI :
10.1109/SECON.1996.510101
