Model-Free Control Approach for Fixed-Wing UAVs with Uncertain Parameters Analysis

Abstract

This paper presents first results of an innovative Model-Free Control (MFC) architecture applied to fixed-wing UAVs. MFC is an algorithm dedicated to systems with poor modeling knowledge. Indeed, the costs to derive a reliable and representative aerodynamic model for UAVs motivated the use of such a controller. By exploiting a purely numerical model, this algorithm provides an intuitive method to tune the control loop without any information about the controlled system. We propose to extend the MFC architecture to the case of fixed-wing UAVs and study the MFC properties in terms of uncertain parameters. As a first result, our designed MFC architecture provides a continuous controller able to stabilize the entire flight envelope of two different fixed-wing UAVs. These results show promising adaptive perspectives and demonstrate that MFC presents robust properties for both uncertain parameters and disturbance rejection.