Enabling an unmanned aircraft system (UAS) to follow swift changes in the waypoints of a trajectory is a challenging task when performed in real-time. By employing a dynamic system model and solving a finite-horizon optimal control problem, Model Predictive Control (MPC) emerges as an effective potential solution for this scenario. However, the actual model of a UAS is rarely precisely known, even more if the UAS is a commercial off-the-shelf platform with internal controllers running onboard an embedded autopilot. To address this challenge, a UAS model identification procedure is performed by collecting input/output data from experimental flights. The obtained dynamic model of the UAS is utilized for the real-time implementation of an MPC for translational motion stabilization. To validate our approach, real-world experiments are conducted to make translational velocities close to the physical constraints of the UAS. The MPC strategy is compared to a proportional–integral–derivative (PID). Results indicate the superiority of the proposed MPC in terms of accuracy and control performance.
Real-time Implementation of Model Predictive Control for Fast Trajectory Tracking of a Quad Rotorcraft UAS
11.12.2023
214453 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Springer Verlag | 2005
|Quad rotorcraft control : vision-based hovering and navigation
TIBKAT | 2013
|Modeling the Quad-Rotor Mini-Rotorcraft
Springer Verlag | 2013
|Real-Time Tracking of Quad-Marked Surfaces
British Library Conference Proceedings | 2002
|