Self-triggered model predictive attitude tracking control for quadrotor UAVs

Abstract

This paper proposes a horizon-shrinking self-triggered model predictive control (HSSTMPC) strategy for the attitude tracking control of a class of perturbed quadrotor unmanned aerial vehicle (UAV) systems. By jointly designing a selftriggering mechanism and a horizon-shrinking strategy, the solving frequency and dimension of the optimal control problem are significantly reduced, thereby alleviating the overall computational burden. Furthermore, incremental input constraints are incorporated alongside standard state and control input constraints. This fundamentally mitigates the risk of severe rotor thrust variations, reducing mechanical wear and enhancing flight safety. Theoretical analyses of feasibility and closed-loop stability rigorously demonstrate the robust convergence of the proposed algorithm. Finally, simulation and experimental results verify that the proposed control scheme enables safe and accurate attitude-tracking control for quadrotor UAVs with high computational resource efficiency.