Reliable spacecraft attitude control depends on accurate prediction of attitude dynamics, particularly when model-based strategies such as Model Predictive Control (MPC) are employed, where performance is limited by the quality of the internal system model. For spacecraft with complex dynamics, obtaining accurate physics-based models can be difficult, time-consuming, or computationally heavy. Learning-based system identification presents a compelling alternative; however, models trained exclusively on data frequently exhibit fragile stability properties and limited extrapolation capability. This work explores Physics-Informed Neural Networks (PINNs) for modeling spacecraft attitude dynamics and contrasts it with a conventional data-driven approach. A comprehensive dataset is generated using high-fidelity numerical simulations, and two learning methodologies are investigated: a purely data-driven pipeline and a physics-regularized approach that incorporates prior knowledge into the optimization process. The results indicate that embedding physical constraints during training leads to substantial improvements in predictive reliability, achieving a 68.17% decrease in mean relative error relative. When deployed within an MPC architecture, the physics-informed models yield superior closed-loop tracking performance and improved robustness to uncertainty. Furthermore, a hybrid control formulation that merges the learned nonlinear dynamics with a nominal linear model enables consistent steady-state convergence and significantly faster response, reducing settling times by 61.52%-76.42% under measurement noise and reaction wheel friction.
Intelligent Control of Spacecraft Reaction Wheel Attitude Using Deep Reinforcement Learning
Swinburne University of Technology（斯winburne大学）；University of South Australia（南澳大利亚大学）；Systems Research Institute Polish Academy of Sciences（波兰科学院系统研究所）
Department of Aeronautics and Astronautics, ITB（印尼万隆理工大学航空航天系）
In general, most of communication satellites were designed to be operated in geostationary orbit. And many of them were designed in prolate dual-spin configuration. As a prolate dual-spin vehicle, they have to be stabilized against their internal energy dissipation effect. Several countries that located in southern hemisphere, has shown interest to use communication satellite. Because of those countries southern latitude, an idea emerged to incline the communication satellite (due to its prolate dualspin configuration) in elliptical orbit. This work is focused on designing Attitude Stability S…
Attitude estimation or determination is a fundamental task for satellites to remain effectively operational. This task is furthermore complicated on small satellites by the limited space and computational power available on-board. This, coupled with a usually low budget, restricts small satellites from using high precision sensors for its especially important task of attitude estimation. On top of this, small satellites, on account of their size and weight, are comparatively more sensitive to environmental or orbital disturbances as compared to their larger counterparts. Magnetic disturbance f…
Convolutional Neural Network Design and Evaluation for Real-Time Multivariate Time Series Fault Detection in Spacecraft Attitude Sensors
Airbus Defence and Space GmbH（阿依斯堡防务与航天有限公司）；Department of Space Systems Engineering, Faculty of Aerospace Engineering, Delft University of Technology（航天系统工程系，航空航天工程学院，代尔夫特理工大学）
Imitation Learning for Satellite Attitude Control under Unknown Perturbations
Mechanical and Aerospace Engineering Department, Rutgers, the State University of New Jersey（机械与航空航天工程系，新泽西州立大学拉特格斯分校）；Aerospace Engineering Department, Embry-Riddle Aeronautical University（航空航天工程系，埃姆布里-瑞德航空大学）
Department of Aeronautics and Astronautics, ITB（万隆理工大学航空航天系）；Department of Aerospace-IT Fusion Engineering, Konkuk University（建国大学航空航天-IT融合工程系）
In response to the interest to re-use Palapa B2R satellite nearing its End of Life (EOL) time, an idea to incline the satellite orbit in order to cover a new region has emerged in the recent years. As a prolate dual-spin vehicle, Palapa B2R has to be stabilized against its internal energy dissipation effect. This work is focused on analyzing the dynamics of the reusable satellite in its inclined orbit. The study discusses in particular the stability of the prolate dual-spin satellite under the effect of perturbed field of gravitation due to the inclination of its elliptical orbit. Palapa B2R p…
Over the past few decades, continuous quaternion-based attitude control has been proven highly effective for driving rotational systems that can be modeled as rigid bodies, such as satellites and drones. However, methods rooted in this approach do not enforce the existence of a unique closed-loop (CL) equilibrium attitude-error quaternion (AEQ); and, for rotational errors about the attitude-error Euler axis larger than πrad, their proportional-control effect diminishes as the system state moves away from the stable equilibrium of the CL rotational dynamics. In this paper, we introduce a new ty…
This paper presents a method for guiding a robot manipulator to capture and bring a tumbling satellite to a state of rest. The proposed approach includes developing a coordination control for the combined system of the space robot and the target satellite, where the satellite acts as the manipulator payload. This control ensures that the robot tracks the optimal path while regulating the attitude of the chase vehicle to a desired value. Two optimal trajectories are then designed for the pre- and post-capture phases. In the pre-capturing phase, the manipulator manoeuvres are optimized by minimi…
This work introduces the Spacecraft Pose Network (SPN) for on-board estimation of the pose, i.e., the relative position and attitude, of a known non-cooperative spacecraft using monocular vision. In contrast to other state-of-the-art pose estimation approaches for spaceborne applications, the SPN method does not require the formulation of hand-engineered features and only requires a single grayscale image to determine the pose of the spacecraft relative to the camera. The SPN method uses a Convolutional Neural Network (CNN) with three branches to solve for the pose. The first branch of the CNN…
