Model Predictive Control of a Tilt-Rotor UAV for Load Transportation

In this paper a model predictive control (MPC) is used to solve the path tracking problem of a tilt-rotor unmanned aerial vehicle while it carries a suspended load. The MPC is designed based on the linear error model of the system, which is linearized around a generic trajectory. The control action is calculated via an optimization problem, where a cost function is solved taking into account input and state constraints. Furthermore, the MPC considers a terminal cost in order to ensure stability, allowing the prediction horizon reduction. The multibody non-linear dynamic model is obtained via Euler-Lagrange formulation, assuming four rigid bodies and ten degrees of freedom (DOF) of the vehicle. Numerical simulations are performed with the objective to evaluate the controller considering constant disturbances at different instants of time, and modeling errors.