Ph. D. Project
Restricted model control of a tilt-body convertible drone.
2023/10/16 - 2026/10/15
Other supervisor(s):
CONDOMINES Jean-Philippe (
During the various projects carried out by ENAC and CRAN, the need arose to think about the development of robust control laws for
so-called "tilt-body" UAVs. This UAV structure offers the possibility to land and take off vertically but also to make a transition to a long
duration flight, at very high speed, which no standard UAV configuration, whether fixed wing or quadrotor, can do today. Convertible
tilt-fuselage drones offer the advantage of not using articulated parts on board, which are a source of weight and fragility. However, the
optimal design of tilt-body drones poses two major challenges. The first concerns the aerodynamic performance of the craft, which
results from a compromise between two very different flight phases: the stationary phase and the forward phase. The UAV must be
carefully sized to find the best compromise between these two phases. The second challenge is to ensure a safe and controlled
transition between the two phases of flight. This last point, which we will address during this thesis, requires a very thorough study of
the robustness and accuracy of the control laws. Indeed, it is not enough to switch from a vertical to a horizontal mode while crossing a
river, as quickly as possible. It is necessary to be able to remain in stable equilibrium in the intermediate phases because the drone
must be able to fly by crosswind. This is the concept of "balanced transition" which is only possible by using sophisticated control laws.

The objective of this phd thesis is twofold. 1) To evaluate in simulation the performances of a control called restricted model combining
flatness and MFC (Model-Free Control) for a convertible drone of type "til-body". Thus, after a bibliographical research stage, the work
will focus on the one hand on the manipulation of an existing model of convertible UAV coded in Matlab/Simulink; and on the other
hand on the implementation and the evaluation of the performances of the control produced allowing to make the UAV navigate. 2) In
real flight, with UAVs developed internally at ENAC, to evaluate the performances of the proposed controls. Several flight configurations
associated with different types of disturbances will be considered.
Control Identification Diagnosis