Closed-loop MIMO data-driven attitude control design for a multirotor UAV

In this paper, the problem of tuning the attitude control system of a multirotor unmanned aerial vehicle (UAV) is tackled and a data-driven approach is proposed. With respect to previous work, the data used to tune the controller gains is collected in flight during closed-loop experiments. Furthermore, the simultaneous tuning of roll and pitch attitude control loops is demonstrated, thus paving the way to MIMO data-driven attitude control design. Simulation results confirmed that a MIMO controller allows rejecting undesired coupling effects that affect the performance of a standard decoupled controller usually employed in autopilots for multirotor UAVs. Finally, the results based on experimental work carried out on a quadrotor UAV show that a good level of performance can be achieved in typical operating conditions with the proposed tuning method.