Metraipolis › archive › dc_4b517cd002c5
GINIO: A Geometric SO(3)-Equivariant Interface for Neural Inertial Odometry — arXiv:2609.25338
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.25338
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T15:08:12.435Z
- SHA-256 of the archived text
ea6d60bd93ed4b7d6b463fa1df5b3986156701285f2d42f5b348734164b5a1f9- Stable URL of this page
https://metraipolis.com/documento/dc_4b517cd002c5
How to cite
GINIO: A Geometric SO(3)-Equivariant Interface for Neural Inertial Odometry — arXiv:2609.25338 — arXiv (en), https://arxiv.org/abs/2609.25338 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4b517cd002c5 (Metraipolis archive dc_4b517cd002c5, sha256 ea6d60bd93ed4b7d6b463fa1df5b3986156701285f2d42f5b348734164b5a1f9)
BibTeX
@misc{dc4b517cd002c5,
title = {GINIO: A Geometric SO(3)-Equivariant Interface for Neural Inertial Odometry — arXiv:2609.25338},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.25338},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4b517cd002c5; sha256 ea6d60bd93ed4b7d6b463fa1df5b3986156701285f2d42f5b348734164b5a1f9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25338. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Neural inertial odometry increasingly uses networks as learned measurements inside filtering pipelines. Such measurements should transform consistently under arbitrary IMU mounting conventions: their mean must transform as a vector, and their covariance must transform congruently as a second-order tensor. We present GINIO, a geometric SO(3)-equivariant interface for neural inertial odometry under arbitrary rotations of the IMU measurement frame. Given calibrated IMU windows, our framework predicts a motion measurement and uncertainty obeying these tensorial laws. To support efficient sensor-frame learning, we introduce Last-Frame Alignment (LFA), a deterministic preprocessing step that is provably equivalent to world-frame training for SO(3)-equivariant predictors. The connected estimator tracks sensor-local states such as IMU bias, separating nuisance estimation from the geometric law enforced by the learned measurement. We instantiate the same interface in filter-connected NIO, AirIO-style recurrent aerial prediction, EqNIO-style full-SO(3) canonicalization, and ResNet-style temporal backbones. On TLIO, GINIO achieves 2.018 m ID/SO(3) ATE while EqNIO degrades to 76.389 m, using 11.6x fewer FLOPs. On NanoBench, our AirIO-style instantiation improves ATE from 5.579 m to 1.430 m without external attitude input, and our ResNet-style instantiation reaches 0.581 m ATE versus 0.645 m for ResNet1D. On Fetch, GINIO empirically reduces unseen physical-remount ATE from 8.15 m to 0.50 m without retraining, demonstrating robustness beyond the exact coordinate-frame guarantee. For uncertainty, spectral covariance reduces covariance-equivariance error by over three orders of magnitude compared with a diagonal head. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.LG). Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.25338 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T14:37:29.521Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).