SIGNN++: Robust Graph Neural Star Identification on Earth
Abstract
We present SIGNN++, a learning-based star identification (SID) framework for celestial navigation in GNSS-denied nighttime environments. Although researchers have developed learning-based SID methods to improve robustness to observation noise, prior evaluations have focused mainly on synthetic or non-terrestrial environments, leaving their reliability under real terrestrial conditions unclear. To address these limitations, SIGNN++ extends SIGNN, a graph neural network-based SID method, with three components: a more realistic noise model for training data, Fourier Transform (FT) of angular-distance edge features, and RANSAC-based Perspective Star Alignment for identification and attitude refinement. The results show that SIGNN++ is more robust to terrestrial observation disturbances and achieves higher star identification accuracy and solved rates than the original SIGNN and publicly available SID baselines, while maintaining comparable attitude accuracy and practical computation time for continuous operation. The implementation code of SIGNN++ and datasets will be available at https://signnpp.github.io/signnpp/.
Related
- Comparative analysis of Neural Networks approaches for Initial Orbit Determination of Near-Earth Objects
- NuGNN: A Graph Neural Network for Nuclear Reaction Network Equations
- The Wavelet Detection Filter: a real time unmodeled pipeline for gravitational wave transients, ranking coincidences with a graph neural network
- Graph Neural Networks for Astrophysics and Cosmology
- Graph neural networks for exoplanet atmospheres
BibTeX
@misc{sohn2026signn,
title = {{SIGNN++: Robust Graph Neural Star Identification on Earth}},
author = {Jeongbin Sohn and Jina Lee and Max Yen and Dowan Gwon and Jeahn Han and Floyd Hepburn-Dickins and Brian Coltin and Ryan Soussan and Pyojin Kim},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.30186},
eprint = {2610.30186},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.30186},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - SIGNN++: Robust Graph Neural Star Identification on Earth
AU - Jeongbin Sohn
AU - Jina Lee
AU - Max Yen
AU - Dowan Gwon
AU - Jeahn Han
AU - Floyd Hepburn-Dickins
AU - Brian Coltin
AU - Ryan Soussan
AU - Pyojin Kim
PY - 2026
DA - 2026/10/20
PB - arXiv
DO - 10.48550/arxiv.2610.30186
UR - https://arxiv.org/abs/2610.30186
AB - We present SIGNN++, a learning-based star identification (SID) framework for celestial navigation in GNSS-denied nighttime environments. Although researchers have developed learning-based SID methods to improve robustness to observation noise, prior evaluations have focused mainly on synthetic or non-terrestrial environments, leaving their reliability under real terrestrial conditions unclear. To address these limitations, SIGNN++ extends SIGNN, a graph neural network-based SID method, with three components: a more realistic noise model for training data, Fourier Transform (FT) of angular-distance edge features, and RANSAC-based Perspective Star Alignment for identification and attitude refinement. The results show that SIGNN++ is more robust to terrestrial observation disturbances and achieves higher star identification accuracy and solved rates than the original SIGNN and publicly available SID baselines, while maintaining comparable attitude accuracy and practical computation time for continuous operation. The implementation code of SIGNN++ and datasets will be available at https://signnpp.github.io/signnpp/.
ER -