Control of unknown nonlinear systems using output feedback

Kosmatopoulos et al. (1999) have introduced a switching adaptive/neural control algorithm for the state feedback stabilization of completely unknown nonlinear systems. In this paper, we address the problem of output feedback regulation/stabilization of unknown nonlinear systems. We develop and analyze a controller which, using output feedback information, can regulate/stabilize any nonlinear system that satisfies: 1) the system is state feedback stabilizable, 2) it has a bounded input/output bounded state property, and 3) its vector-fields are analytic. The proposed controller consists of the switching adaptive/neural controller used for state feedback stabilization combined appropriately with Volterra/Wiener neural networks. The Volterra/Wiener neural network uses the system input/output information for the estimation of unknown, state-dependent functions. The estimates of these functions are used by a switching adaptive mechanism for the regulation/stabilization of the system