In this paper, a neural network based adaptive controller is designed for a class of nonlinear systems. The offline neural network training and on-line neural network tuning are integrated to assure that not only the stability of the resulting closed-loop control system is guaranteed, but also the reasonable tracking performance is achieved. The adaptation of the parameters of neural networks is handled based on the robust adaptive control design methodology. The off-line training step incurs additional cost and maybe inconvenience compared to direct on-line neural network parameters tuning. However, the stability analysis and performance evaluation have a more solid basis; and the weight adaptation laws are different than those existing in the literature and bear more practical meaning and significance. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Chen, D., & Yang, J. (2005). Stability analysis and performance evaluation of an adaptive neural controller. In Lecture Notes in Computer Science (Vol. 3498, pp. 42–47). Springer Verlag. https://doi.org/10.1007/11427469_7
Mendeley helps you to discover research relevant for your work.