Gaussian Process Regression-based Disturbance Rejection Controller for Unmanned Aerial Vehicle

Abstract

This paper proposes a disturbance rejection controller for an unmanned aerial vehicle (UAV) in the presence of the model uncertainties and unknown external disturbances. The main objective of this study is an investigation of the feasibility of using the Gaussian process regression (GPR), which is one of the data-driven approaches, as a disturbance estimator. First, the baseline controller is designed by the two-loop feedback linearization control. The GPR is then augmented in the baseline controller for estimating and compensating for the disturbances. The simulations are performed to validate the performance of the proposed control structure.