Hybrid Model Predictive Control with Physics-Informed Neural Network for Satellite Attitude Control

TL;DR

The paper tackles reliable satellite attitude control under nonlinear dynamics and disturbances by integrating Physics-Informed Neural Networks (PINNs) into a hybrid Model Predictive Control (MPC) framework. A multilayer perceptron learns the attitude dynamics with a physics-based loss, balanced against data-driven loss via a Lagrangian dual, achieving a 68.17% reduction in 10-step mean relative error. The learned dynamics feed a nonlinear MPC, which can switch to a linear state-space model when the attitude error falls below $1^\circ$, yielding substantially faster convergence (settling times down to $43.9$ s) and near-zero steady-state error under disturbances and friction. Results demonstrate improved predictive reliability and closed-loop robustness, highlighting the potential of hybrid data-driven and physics-informed methods for high-stakes spaceflight control tasks.

Abstract

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.

Figures (2)

• Figure 1: Schematic of the proposed hybrid MPC with learned dynamics. When the attitude error is below 1 degree we switch from a nonlinear MPC with PINN state estimator (red) to a linear MPC (yellow).
• Figure 2: 300 MC simulation with parameter estimation errors, state estimation noise and RWs friction for MLP-LD (left), traditional MPC with linear (center) and MLP-LD + Linear (right).