Abstract

Over the past few decades, there has been a significant need for the use of Uninhabited Aerial Vehicles (UAVs) in both civilian and military applications. Nowadays, UAVs are becoming increasingly popular to perform hard missions that include surveillance, sensing (of, for example, chemical agents), acquiring weather data, and reconnaissance over oceanic and remote areas. Motivated by such importance of UAV missions, this thesis presents a new control methodology applied to a UAV path following problem in the longitudinal plane. This control methodology takes into account the actuator dynamics that is used to deflect the elevator of the UAV. In particular, it considers a model of the Coulomb friction that exists in the dynamics of the actuator. For controller design purposes, the overall dynamics of the UAV are divided into two sets of dynamics that are in cascade connection. One set of dynamics describes the steering motion of the UAV and another set describes the translational motion of the UAV, where both motions are in the longitudinal plane. Each set is treated separately in the controller design. A piecewise-affine state feedback controller is designed for the dynamics of the steering subsystem of the UAV and a nonlinear controller is designed for the dynamics of the translational velocity subsystem of the UAV. Stability of the novel cascade interconnection of the two subsystems is investigated. Simulation results show the effectiveness of the proposed control methodology.