A planar path-following model predictive controller for fixed-wing Unmanned Aerial Vehicles

This paper addresses the design of a path-following using a sampled-data Model Predictive Control (MPC) approach for a fixed-wing Unmanned Aerial Vehicle (UAV). In path-following, the motion task consists in steering the vehicle towards a given desired geometric path and make it follow along the path with an assigned speed profile. To this end, we propose an MPC controller that will generate desired linear and angular velocities using a simplified kinematic model of a planar UAV. The output of the MPC is then fed to a low-level controller (the autopilot) that is tasked to track such reference signals. This combination allows the design of the MPC controller for a subset of the state variables, with the consequent reduction of computational burden, and to explicitly handle input constraints. Simulation results show the performance of the proposed scheme.