OPEN-LOOP SUBSPACE IDENTIFICATION OF A FLEXIBLE UNMANNED AERIAL SYSTEM

Abstract

This paper presents the identification of an Unmanned Aerial System (UAS) with flexible wings from open-loop data using subspace methods. For aerodynamic and flight control systems, a reliable model is important to comprehend the system behaviour and to design a feedback loop, that can be applied as well to minimize the effects of structural flexibility. So, a parametric model identification for flexible aircraft applying subspace techniques was performed. Preliminary results presented in this paper are related to identification using synthetic data. Finally, it is shown the experimental results from the first flight test performed in open-loop operation. The experimental results reveals that subspace methods estimate a state-space model suitable, with better fit for the range of frequencies of the experimental data. Therefore, it was not possible to obtain a representative model for a broader frequency range, however, this is not a limitation of the method, but a persistent excitation problem associated with the limited frequencies in the input signal.