Fixed-wing UAV with transitioning flight capabilities : Model-Based or Model-Free Control approach? A preliminary study

Abstract

Transitioning vehicles experience three different flight phases during typical missions. The hovering and forward flight phases have been researched widely, however the transition phase in between is more challenging and has been the subject of less research. One of the control approaches to handle the transition phase relies on model-based methods which require sophisticated wind-tunnel characterization. Accurate modeling of force and moments of a partially stalled wing and control surfaces is highly challenging and time consuming. In addition, these models usually require several flight measurements (such as angle of attack and low airspeed) that are difficult to obtain. As an alternative, some control approaches manage the transition phase without the need for sophisticated models. One example of such an approach is the Model Free Control (MFC). This paper compares the results obtained from both MFC and Linear Quadratic Regulator (LQR) applied to fixed-wing UAV with transitioning flight capability during hovering, transition and forward flight modes. Both of the controllers are designed for a transitioning vehicle called MAVion. The simulation results demonstrated that MFC increases the stability of the aircraft, especially in disturbed flight conditions.