Commande d'un minidrone à hélice carénée : de la stabilisation dans le vent à la navigation autonome
Jean-Michel Pflimlin · HAL (Le Centre pour la Communication Scientifique Directe) · 2006
The design of autonomous navigation strategies for ducted fan unmanned aerial vehicles has now become an important research area. Those small vehicles, able to perform stationary flight, are of great interest for military and civilian applications. The robustness to wind perturbations is the main challenge of those vehicles. Due to the low Reynolds number and the atypical shape of the body, the characterization of the aerodynamic forces applied to ducted fan rotorcrafts is still a difficult problem. Therefore, it is important to estimate those efforts on-line in order to counter them in control. After a modelling step, we extract a cascade structure of the system in order to perform nonlinear control design based on backstepping techniques. We propose a hierarchical controller separated in a high-level position control and a low level attitude control. Before implementing this control on the vehicle, we have to retrieve the position and attitude from the measurements of the sensors. We develop nonlinear filtering techniques on the special orthogonal group to estimate attitude. We also perform IMU/GPS hybridazition to estimate the position. The estimated state is used in the feedback control to ensure hovering flight. In a last part, we complete the control scheme to perform autonomous navigation of the vehicle amidst obstacles. The efficiency of the method is illustrated by simulations and experiments led on the ducted fan VTOL UAV HoverEye developped by Bertin Technologies.