Mixed-Integer vs. Continuous Model Predictive Control for Binary Thrusters: A Comparative Study
2026
Read Mixed-Integer vs. Continuous Model Predictive Control for Binary Thrusters: A Comparative Study (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-19796.txt.
Introduction
Stark, Middelberg, and Vyas compare control methods for spacecraft with binary on-off thrusters. Such thrusters pose a problem for standard continuous controllers. One route solves a mixed-integer control problem directly. The other converts continuous output into binary commands by modulation. The authors add a third variant informed by the modulator state. All three ran in simulations on ESA's REACSA platform. The mixed-integer route used fuel best at low thrust. The plain modulated route grew unstable at higher thrust. Their informed variant closed much of the gap.
Catalog record
| YEAR | 2026 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36064 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.19796
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Mixed-Integer vs. Continuous Model Predictive Control for Binary Thrusters: A Comparative Study Abstract Binary on/off thrusters are commonly used for spacecraft attitude and position control during proximity operations. However, their discrete nature poses challenges for conventional continuous control methods. The control of these discrete actuators is either explicitly formulated as a mixed-integer optimization problem or handled in a two-layer approach, where a continuous controller’s output is converted to binary commands using analog-to-digital modulation techniques such as Delta-Sigma ( Δ Σ \Delta\Sigma ) -modulation. This paper provides the first systematic comparison between these two paradigms for binary thruster control, contrasting continuous Model Predictive Control (MPC) with Delta-Sigma modulation against direct Mixed-Integer MPC (MIMPC) approaches. Furthermore, we propose a new variant of MPC for binary actuated systems, which is informed using the state of the Delta-Sigma Modulator. The two variations for the continuous MPC along with the MIMPC are evaluated through extensive simulations using ESA’s REcap-ACrobat-SAtsim (REACSA) platform. Results demonstrate that while all approaches perform similarly in high-thrust regimes, MIMPC achieves superior fuel efficiency in low-thrust conditions. Continuous MPC with modulation shows instabilities at higher thrust levels, while binary informed MPC, which incorporates modulator dynamics, improves robustness and reduces the efficiency gap to the MIMPC. It can be seen from the simulated and real-system experiments that MIMPC offers complete stability and fuel efficiency benefits, particularly for resource-constrained missions, while continuous control methods remain attractive for computationally limited applications. \papernumber 086 keywords model predictive control, mixed-integer optimization, satellite control {authorList} 4cm \addAuthor Franek Stark Robotics Researcher, Robotics Innovation Center, German Research Center for Artificial Intelligence GmbH \rorlink https://ror.org/01ayc5b57, Bremen, Germany. \emailAddress [email protected] \addAuthor Jakob Middelberg Research Assistant, Robotics Innovation Center, German Research Center for Artificial Intelligence GmbH \rorlink https://ror.org/01ayc5b57, Bremen, Germany. \emailAddress [email protected] Master Student, University of Bremen \rorlink https://ror.org/04ers2y35, Germany. \emailAddress [email protected] \addAuthor Shubham Vyas Robotics Researcher, Robotics Innovation Center, German Research Center for Artificial Intelligence GmbH \rorlink https://ror.org/01ayc5b57, Bremen, Germany. \emailAddress [email protected] Nomenclature \theta = Platform orientation (yaw) in world coordinates \dot{\theta} = Platform angular velocity in world coordinates x,y = Platform position in world coordinates \dot{x},\dot{y} = Platform velocity in world coordinates \omega_{RW} = Reaction Wheel (RW) rotational speed \mathrm{s}_{\theta},\mathrm{c}_{\theta} = Sine and cosine of \theta \textbf{x},\textbf{u} = State and input vector of the system \hat{\textbf{x}} = Target state F_{\text{n}} = Nominal thrust force applied by a single thruster r = Platform radius m,I_{\text{S}} = Platform overall mass and inertia on the z -axis. I_{\text{RW}} = RW ’s inertia on the z -axis. \textbf{x}_{j|t},\textbf{u}_{j|t} = State and input prediction of time step j , predicted at time step t N = Prediction Horizon \Delta\mathrm{t} = Discretization time of the system dynamics within the controller \textbf{U}_{t},\textbf{X}_{t} = Set of all predicted states and inputs at time step t \textbf{u}_{\text{bin},t} = Set of all predicted binary inputs at time step t \mathbf{Q},\mathbf{W} = Diagonal cost matrices for state and input cost terms. 1 Introduction Satellites and space vehicles are commonly equipped with either cold gas or hypergolic thrusters for attitude and position control during proximity operations [ 1 ] . These thrusters are usually binary actuated, i.e., they are on/off devices and can only be either fully on or fully off. This is due to the fact that the thrusters are ordinarily designed to operate at a specific pressure and flow rate, and operating them at partial thrust can lead to instability and inefficiency. Furthermore, on/off thrusters are often simpler than variable-thrust thrusters, leading to higher reliability and lower cost. However, most control methods assume a smooth and continuous state and control input, which is not given with on/off thrusters. Therefore, many existing control methods outlined in the literature are not directly applicable, making it challenging to achieve precise attitude and position control. This results in two options for the development of control systems for proximity operations with binary thrusters. The first is to use a two-layer control architecture, where a continuous controller generates the desired control inputs, and a separate module converts these inputs into binary commands for the thrusters. Different methods have been proposed, from using simple thresholding logic [ 2 ] , up to modulators like pulse-width modulation (PWM) [ 3 ] or \Delta\Sigma -modulation [ 4 , 5 ] or similar schemes to convert analog/continuous control input to digital/discrete values. The second option is to directly perform control on the binary control inputs using Mixed Integer (MI) methods. Until recently, these were considered too computationally expensive for real-time control, but recent advances in MI optimization have made it possible to solve these problems in real-time [ 6 , 7 , 8 , 9 ] . The first approach, while computationally advantageous, can lead to suboptimal performance, as the continuous controller may not be able to fully account for the limitations of the binary thrusters, such as dwell-time constraints (ramp-up or cool-down time). The second approach, while more complex, can lead to better performance, as it can directly account for the limitations of the binary thrusters. While earlier work has compared the performance of PWM and \Delta\Sigma -modulation [ 4 ] , to the best of the authors’ knowledge, there has been no work comparing the performance of the two-layer approach with MI methods for controlling systems with binary thrusters. Such a comparison would allow for understanding the trade-offs between the two approaches and help in understanding if the increased effort for MI methods is justified by the performance gains. In this paper, we compare the performance of a two-layer control architecture using a continuous input Model Predictive Control (MPC) with \Delta\Sigma -modulation with a Mixed Integer Model Predictive Control (MIMPC) approach for controlling a system with on/off thrusters. We also introduce a new MPC approach which uses a continuous input MPC , that is informed about the behavior of the \Delta\Sigma -modulator in the prediction horizon. This allows the MPC to account for the limitations of the binary thrusters while still being able to use efficient numerical methods for solving the continuous optimization problem. We compare the performance of these three approaches in terms of their ability to reach and maintain a target position, as well as their efficiency in terms of thruster usage. The comparison is performed through a simulation study along with experimental validation of the newly proposed binary-informed MPC approach at the Orbital Robotics & GNC Lab (ORGL) of the European Space Agency (ESA) [ 10 ] . Both the simulation and hardware datasets 1 1 1 Datasets: https://doi.org/10.5281/zenodo.18454916 , together with the controller implementation 2 2 2 Controller implementation: https://github.com/dfki-ric-underactuated-lab/mimpc , are publicly available as open source resources. 2 System Description Figure 1 : Sketch (left) of the free-floating platform REACSA (right) The system, which serves as an example for a satellite or small spacecraft, is ESA ’s free-floating platform REACSA [ 11 ] . It is a 200\text{\,}\mathrm{kg} heavy air-bearing platform with a propulsion system containing eight cold-gas thrusters and an additional Reaction Wheel (RW) . A sketch of the system, showing the thruster configuration, is shown along with a picture in Figure 1 . The cold-gas propulsion enforces a minimum thruster activation time t_{\text{on, min}}=$0.1\text{\,}\mathrm{s}$ and limits the maximum firing time of a single thruster to t_{\text{on, max}}=$0.3\text{\,}\mathrm{s}$ . Further, it