MPC for underactuated spacecraft control with a Lyapunov supervised physics-informed neural network correction layer

2026-06-11

AI Summary

This work addresses the challenge of achieving high-precision attitude stabilization for underactuated spacecraft, whose limited control degrees of freedom and susceptibility to disturbances hinder conventional methods. The authors propose a three-layer hierarchical control architecture: a top layer employing nonlinear model predictive control (NMPC) for constraint-aware attitude trajectory planning; a middle layer utilizing an offline-trained physics-informed neural network (PINN) to estimate unmodeled disturbance torques; and a bottom layer incorporating a Lyapunov-based supervisory mechanism that online constrains learning-based correction terms to guarantee closed-loop stability. This approach innovatively integrates data-driven disturbance compensation with rigorous stability assurance, enabling graceful degradation to pure model-based control if the learning module fails. Monte Carlo simulations demonstrate significantly reduced steady-state attitude errors compared to baseline NMPC, while maintaining robustness and safety under inertia uncertainties and external disturbances.

Abstract

Underactuated spacecraft faces controllability limitations and heightened sensitivity to environmental disturbances, complicating attitude maneuvering and stabilization. Due to the lack of control authority along the underactuated axis, conventional controllers cannot directly stabilize all attitude components and therefore require reference planning strategies. Furthermore, MPC approaches remain sensitive to inertia uncertainty and unmodeled dynamic couplings, resulting in degraded tracking performance under mismatch. To address these issues, we consider a hierarchical architecture integrating three layers: (i) a nonlinear model predictive controller (NMPC) for constraint and underactuation-aware maneuver planning and nominal closed-loop stability under actuator limits; (ii) a physics-informed neural network (PINN) trained offline on simulation data to estimate residual disturbance torques, with loss terms that enforce consistency with rigid-body rotational dynamics; (iii) a Lyapunov-based supervisory safety mechanism that evaluates the learned correction online and bounds or suppresses its influence to preserve the stability properties of the baseline controller. The architecture is evaluated in a high-fidelity simulation environment modelling reaction wheel dynamics, actuator saturation, and environmental disturbances. Monte Carlo studies show statistically significant reductions in steady-state attitude error relative to standalone NMPC while maintaining robust behavior under uncertainty. The supervisory layer ensures graceful degradation to purely model-based control when the learning-based augmentation is unreliable.

Problem

• underactuated spacecraft
• attitude control
• model predictive control
• inertia uncertainty
• disturbance rejection

Innovation

• Model Predictive Control
• Physics-Informed Neural Network
• Lyapunov-based Supervision
• Underactuated Spacecraft
• Hierarchical Control Architecture