Synthesis of hierarchical neural controller for nonlinear systems

The theoretical study on synthesis of hierarchical neural controllers for nonlinear systems affine in control is presented. We first show that performance criteria based optimal neural controllers can be synthesized to approximately identify the switching manifold for control. We then show that the hierarchical neural controller can deal with system uncertainties in parameters which are fixed but unknown, and should perform reasonably well in theory. Further, the adaptive hierarchical neural controllers are developed to deal with systems uncertainties in parameters which are time varying, and it is shown that they are able to perform satisfactorily