Novel modeling technology for UAS-S4 and UAS-S45 flight dynamics

Abstract

The rapid development of unmanned flight systems in the aviation domain and the growing interest in environmental preservation are driving aircraft manufacturers to focus more and more on performance improvement techniques. In this case, improving the performance consists in achieving an optimal flight while reducing the amount of fuel consumed. However, the validation of a performance improvement technique can be very demanding in terms of money and time and can also lead to the destruction of the aircraft. To respond to these problems of financial and environmental resource reduction for the validation of performance improvement techniques and aircraft certification, this thesis proposes the design of a high precision flight simulation model. The main objectives are: 1) to design a dynamic flight model capable of accurately estimating the behaviour of UAS-S4 and UAS-S45 manufactured by Hydra Technologies; and 2) to ensure that the flight dynamics model allows the testing of performance improvement techniques and the implementation of a new control law. This thesis relates essentially to the modeling of the flight dynamics of aircraft as well as to aircraft stability and control. It is articulated around four contributions. The first contribution focuses on the methodologies for obtaining the UAS-S4 and the UAS-S45 flight models. The proposed method divides the overall architecture of each aircraft into sub-models. Subsequently, each of the sub-models is estimated using numerical and experimental methods. This contribution shows that the proposed methodology requires only a minimum of data for the flight dynamics calculations. The second contribution is related to the use of numerical methods for estimating the aerodynamic model of the UAS-S4 and the UAS-S45. In contrast to the various aerodynamic calculation methods available, the proposed method, which is based on nonlinear vortex lattices, provides high-fidelity results in an affordable time. To verify these results, experimental analyses were carried out on a model of the UAS-S45 wing in the Price-Païdoussis subsonic wind tunnel. The third contribution examines the accuracy of fluid dynamics analysis methods as well as an improved blade elements theory for estimating aerodynamic propeller performance of the UAS-S4 and the UAS-S45. The two proposed numerical methods were validated using an experimental study in an open test section wind tunnel. The fourth contribution proposes a new control method, using an LQR approach and a PI with reference feedforward controller for the stability and control of the UAS-S45. The robustness of the system is ensured using to an extended state observer. A fuzzy neural network was subsequently used for gain scheduling.

Date	14 Jan 2020
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Ruxandra Botez (Supervisor)