EUVIMEDEUVIMED entdecken
Crossref

Transfer Alignment for Master–Slave Inertial Navigation Systems Using Factor Graph Optimization

Feng Liu, Bo Yang, Qijia Zhao, Shaolin Lü, Rong Zhang

IEEE Transactions on Instrumentation and Measurement · 2026 · 75 · 8510415-8510415

DOI: 10.1109/tim.2026.3705042 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Instrumentation and Measurement, 2026. Verfasst von Feng Liu und 4 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transfer alignment is a fundamental initialization procedure in master–slave inertial navigation system (INS) architectures, particularly for systems employing low-accuracy inertial sensors. The conventional transfer alignment methods are typically based on recursive filtering with small-angle linearization, which restricts their applicability under large misalignment angles or degraded inertial sensing conditions. This article proposes a factor graph optimization (FGO)-based transfer alignment framework, in which the alignment problem is formulated as a maximum a posteriori (MAP) estimation over a sliding window. High-rate inertial measurements are incorporated using an inertial measurement unit (IMU) preintegration model, while master INS information is introduced through attitude, velocity, and position observation factors with explicit lever-arm compensation. To overcome the limitations imposed by small-angle assumptions, a Lie group-based attitude matching factor is developed directly on the special orthogonal group $\mathrm {SO}(3)$ , enabling geometrically consistent residual modeling that remains valid under both small and large initial misalignment conditions. A fixed-lag smoothing strategy with marginalization is employed to bound computational complexity and support online implementation. The experimental results on both public automotive datasets and real-vehicle field tests demonstrate that the proposed method achieves an average attitude root mean square error (RMSE) reduction of 60.5% relative to a conventional extended Kalman filter (EKF)-based approach across different sensor configurations, while maintaining stable convergence even under large misalignment angles.

Zitieren

APA

Liu, F., Yang, B., Zhao, Q., Lü, S., & Zhang, R. (2026). Transfer Alignment for Master–Slave Inertial Navigation Systems Using Factor Graph Optimization. IEEE Transactions on Instrumentation and Measurement, 75, 8510415-8510415. https://doi.org/10.1109/tim.2026.3705042

Vancouver

Liu F, Yang B, Zhao Q, Lü S, Zhang R. Transfer Alignment for Master–Slave Inertial Navigation Systems Using Factor Graph Optimization. IEEE Transactions on Instrumentation and Measurement. 2026;75:8510415-8510415. doi:10.1109/tim.2026.3705042

BibTeX
@article{Liu2026,
  author = {Liu, Feng and Yang, Bo and Zhao, Qijia and Lü, Shaolin and Zhang, Rong},
  title = {Transfer Alignment for Master–Slave Inertial Navigation Systems Using Factor Graph Optimization},
  journal = {IEEE Transactions on Instrumentation and Measurement},
  year = {2026},
  volume = {75},
  pages = {8510415--8510415},
  doi = {10.1109/tim.2026.3705042}
}
Kostenlos starten →