Adaptive optimal control of a class of uncertain nonlinear systems using fuzzy logic and neural networks

A study on the synthesis of neural network and fuzzy logic based controllers for optimally controlling uncertain nonlinear systems linear in control is presented in this paper. Three different kinds of hierarchical controller architectures are proposed, which include a hierarchical neuro-fuzzy controller architecture, a hierarchical fuzzy-neuro controller architecture, and a hierarchical fuzzy logic controller architecture. This study concludes that the proposed neural-network and fuzzy logic based control schemes are useful for nonlinear system applications. It first shows that fuzzy controllers, besides neural controllers, can be synthesized to approximately identify the switching manifold for optimal control. It then shows that the neuro-fuzzy controller, fuzzy-neuro controller, and hierarchical fuzzy controller can deal with system parametric uncertainties. Further, the adaptive neuro-fuzzy controllers, fuzzy-neuro controllers, and hierarchical fuzzy controllers are developed to deal with systems with time varying parametric uncertainties.