Adaptive output feedback control of an underactuated three degrees-of-freedom model helicopter using neural networks

In this paper we apply a recently developed neural network adaptive output feedback control methodology, suited to a class of nonlinear underactuated dynamical systems, to a three degrees-of-freedom (3DOF) model helicopter. Experimental results are presented and illustrate how this approach performs in the presence of modelling errors and parameter variations. Important topics concerning practical implementation are discussed. The obtained results are promising and show that this technique has good performance when dealing with a real uncertain nonlinear underactuated dynamical system.