Output feedback control for constrained pure-feedback systems: A non-recursive and transformational observer based approach

Abstract

In this paper we investigate the control problem of uncertain pure-feedback systems under time-varying output constraints using output information only. By making use of the salient cascade properties of pure-feedback systems as well as a novel scaling function, we convert the constrained system into a normal form without constraints. Then by using only one single neural network (NN) unit for nonlinear approximation and one high-gain observer for transformed state estimation, an adaptive NN output feedback control scheme is constructed. Different from existing results in the literature, our method exhibits the following features: (1) achieving semi-global stable control with only output feedback without imposing any additional restrictive condition; (2) avoiding the recursive design procedures required by some typical approaches such as backstepping; and (3) recovering the steady-state tracking performance under the state feedback. Besides, all the signals in the closed-loop are bounded and the output constraints are never violated. The effectiveness and flexibility of the developed method is demonstrated through control design and simulation on the non-trivial aircraft short-period dynamics.