Influence of number of measured states on accuracy and robustness of flight control system

Abstract

Recently in many countries the efforts of researchers are directed to the design of low cost, accurate and robust controllers for small Unmanned Aerial Vehicle (UAV) using the minimum number of on-board sensors. The basic problem of such kind of design is to restore the unknown (unmeasured) states. The approach proposed in the paper consists of two procedures: state restoration by the reduced-order observer (Luenberger filter) and optimal deterministic controller design. This technique was applied to the “benchmark” model of UAV longitudinal motion (there were considered two cases: with three and two on-board sensors only) and results of its application are presented.