SER-MPC: Singularity-Free Robust Event-Triggered MPC for UAVs Subject to External Disturbances

This article investigates the trajectory tracking control problem for quadrotor uncrewed aerial vehicles (UAVs) subject to physical constraints and external disturbances. To this end, a position-loop nonlinear model predictive control (MPC) with explicit constraint handling and an attitude-loop controller based on a finite-time disturbance observer are integrated into a two-layer cascaded architecture. To explicitly address the algorithmic singularities inherent in position-attitude cascaded decoupling, we develop a safety-guaranteed decoupling mechanism for the UAV cascaded architecture. By rigorously formulating a singularity-free constraint set for the position loop, the proposed strategy guarantees that the extracted attitude commands are always mathematically well-defined and physically executable. Then, the attitude-loop controller ensures fast and robust attitude tracking in the presence of external disturbances. To further enhance robustness and efficiency, a high-order disturbance observer is incorporated to provide smooth disturbance estimates, and an eventtriggered mechanism (ETM) is introduced to reduce unnecessary online MPC updates. Comparative numerical simulations and real-world flight experiments are carried out, and the obtained results reveal that the proposed framework achieves more reliable singularity-free constraint satisfaction and more accurate trajectory and attitude tracking than the considered baseline methods, while significantly reducing unnecessary online optimizations.