Machine Learning Argument of Latitude Error Model for LEO Satellite Orbit and Covariance Correction

2026-02-18

AI Summary

This study addresses the rapid growth of along-track error in low Earth orbit (LEO) satellite orbit prediction, primarily driven by atmospheric drag modeling inaccuracies, which violates the Gaussian assumption of the state covariance. To mitigate this, the authors propose a novel machine learning–based correction method that exclusively targets the dominant error dimension—the argument of latitude—without modifying the existing orbit propagator. By leveraging single-epoch vector covariance and backward-propagated errors, the approach employs a time-conditioned neural network combined with Gaussian process regression to model and correct unmodeled atmospheric drag effects. The method effectively preserves the physical propagation characteristics in all other state dimensions while significantly improving prediction accuracy, restoring the Gaussianity of the covariance, and successfully extending the validity duration of VCM ephemerides.

Technology Category

Machine Learning: Calibration & Uncertainty QuantificationIntelligent Robots: State EstimationCognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsUser Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systemsGraph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Low Earth orbit (LEO) satellites are leveraged to support new position, navigation, and timing (PNT) service alternatives to GNSS. These alternatives require accurate propagation of satellite position and velocity with a realistic quantification of uncertainty. It is commonly assumed that the propagated uncertainty distribution is Gaussian; however, the validity of this assumption can be quickly compromised by the mismodeling of atmospheric drag. We develop a machine learning approach that corrects error growth in the argument of latitude for a diverse set of LEO satellites. The improved orbit propagation accuracy extends the applicability of the Gaussian assumption and modeling of the errors with a corrected mean and covariance. We compare the performance of a time-conditioned neural network and a Gaussian Process on datasets computed with an open source orbit propagator and publicly available Vector Covariance Message (VCM) ephemerides. The learned models predict the argument of latitude error as a Gaussian distribution given parameters from a single VCM epoch and reverse propagation errors. We show that this one-dimensional model captures the effect of mismodeled drag, which can be mapped to the Cartesian state space. The correction method only updates information along the dimensions of dominant error growth, while maintaining the physics-based propagation of VCM covariance in the remaining dimensions. We therefore extend the utility of VCM ephemerides to longer time horizons without modifying the functionality of the existing propagator.

Problem

• LEO satellite
• orbit error
• atmospheric drag
• covariance correction
• Gaussian assumption

Innovation

• machine learning
• orbit propagation
• covariance correction
• argument of latitude
• VCM ephemerides