Control of uncertain systems by feedback linearization with neural networks augmentation. Part I. Controller design

Abstract

The paper highlights the main steps of adaptive output feedback control for non-affine uncertain systems – both in parameters and dynamics – having a known relative degree. Given a reference model, the objective is to design a controller that forces the measured system output to track the reference model output with bounded errors. A single hidden layer neural network is used to counteract feedback linearization error. A dynamic observer of tracking error is added. The treatment of control saturation is also sketched. The mathematical model for the longitudinal dynamics of an experimental helicopter is used as framework.