Comments on System identification of an UAS model using a subspace method

The aim of this paper is providing an experimental procedure of system identification of an Unmanned Aerial System (UAS) with fixed flexible wing using subspace identification techniques for dynamic systems. A reliable model is important for aerodynamic and flight control systems to comprehend the behavior of the system and to design a feedback loop that can be applied to minimize the effects of structural flexibility. In this scenario, this research is intended to identify a parametric model for a flexible aircraft from open-loop experimental data applying the DSR e algorithm. A flight test campaign was executed using the EOLO, a single engine aircraft with a wingspan of 4 m, weighing in total 8.87 kg. First, results related to identification using synthetic data are presented, in which preliminary estimated parameters based on Ground Vibration Test (GVT) were useful to validate the identified model. Then, the experimental results performed in open-loop operation reveal that subspace algorithms estimate a suitable state-space model covering the entire frequency range contained in the experimental data. It is important to highlight that a limiting factor for obtaining a representative model for the frequency range of interest from the collected data is having a persistently exciting condition for the input signals.