Mixed Integer Model Predictive Control for a free-floating platform with binary and continuous actuation
Franek Stark, Shubham Vyas, Georg Schildbach, Frank Kirchner
In Proceedings of the 2024 CEAS EuroGNC conference, (EuroGNC-2024), 11.6.-13.6.2024, Bristol, CEAS, 2024.

Abstract :

This work develops a novel Mixed Integer Model Predictive Control (MIMPC) for European Space Agency (ESA)’s 3-dof free-floating platform which is actuated using on/off-thrusters that are subject to activation time constraints and a Reaction Wheel (RW). It compares a penalty-term, a Linear Complementarity Constraints (LCC), and a Mixed Integer (MI) based formulation to transcribe the on/off thrusters within the optimization problem. A set of linear constraints is presented to enforce the thruster time constraints. Analyses show that under the activation time constraints and real-time requirements, only the MI formulation provides a functional MPC controller. Hence, an MIMPC which directly controls the system’s eight thrusters and RW is developed. Simulated results show that the controller can (sub-) optimally control and stabilize the system in real time for a short enough prediction horizon. By including the thruster’s timing and on/off constraints, the controller is able to exploit the system’s structure to provide efficient control.

Keywords :

Model Predictive Control; Mixed Integer; Binary thrusters; Dwell time constraints; Space robots

Files:

CEAS-GNC-2024-013.pdf

Links:

https://eurognc.ceas.org/archive/EuroGNC2024/pdf/CEAS-GNC-2024-013.pdf


© DFKI GmbH
last updated 28.02.2023