LMI-based design of a Neuro-Adaptive augmentation controller for an Unmanned Aerial Vehicle

This paper presents a practical approach for verifying worst-case tracking performance of neuro-adaptive systems in presence of bounded uncertainties. Boundedness of the tracking error vector within an a-priori specified compact domain is obtained by applying robust invariant set analysis to the uncertain linear plant where the uncertainty and Neural Network (NN) reconstruction error are considered as norm bounded persistent uncertainties. In this framework it was possible to specify worst-case tracking error requirements via a set of LMIs and to systematically verify the specifications using a numerical LMI solver. The presented method was applied to the performance verification of an adaptive augmentation controller for the short term dynamics of an UAV model.