Control Strategies for Unmanned Aerial Vehicles under Parametric Uncertainty and Disturbances: a Comparative Study

摘要

UAVs within the class of Mini Aerial Vehicles (MAVs) are autonomous aircrafts with low inertias that fly at relatively low speeds. In this sense, MAVs are exposed to high airspeed uncertainties since unexpected changes in wind velocity represent an important percentage of the total airspeed of the vehicle. Moreover, such changes directly modify the aerodynamic forces acting along the vehicle, which leads to important variations on their acceleration owed to their low inertia. Although the structure of the dynamic model of an aircraft is well known, important difficulties arise on the identification of an specific MAV due to its particular characteristics. Thus, modelling errors become an additional source of uncertainty when control algorithms are designed. In this situation, studying the ability that different control strategies present in performing trajectory tracking is of great interest on the development of applications for this type of UAVs. In this paper a comparative study of four control strategies is presented. All algorithms have been implemented in a MAV flight computer. Results from both, Hardware-In-the-Loop (HIL) simulations and real flight experiments, are presented as the main contribution of this work.