Pattern-based NN control of a class of nonlinear systems

Abstract

This paper studies the pattern-based neural network (NN) control approach for a class of uncertain nonlinear systems. Firstly, in the identification phase, adaptive NN controllers are designed to achieve closed-loop stability and tracking performance of nonlinear systems for different control situations, and the closed-loop control system dynamics are identified via deterministic learning. The identified dynamics are stored in constant radial basis function (RBF) NNs, and a set of pattern-based constant NN controllers are constructed by using the obtained constant RBF networks. Secondly, still in the phase of identification, when the plant is operated under abnormal conditions but controlled by the normal constant NN controller, the underlying system dynamics are identified via deterministic learning. Thirdly, in the phase of recognition, a bank of estimators is constructed for all the abnormal conditions. When one identified control situation recurs, by using the constructed estimators, the recurred control situation will be rapidly recognized. Finally, in the phase of pattern-based control, the corresponding pattern-based constant NN controller is selected, which guarantees the improved control performance while preserving stability. A simulation example is included to demonstrate the effectiveness of the approach. The results presented in this paper show that pattern-based control may provide a new framework for fast decision and control in dynamic environments.