EUVIMEDEUVIMED entdecken
Crossref

GOF-Based DMI Integrated Solution to Seamless Navigation for Autonomous Vehicles During GNSS Outages

Wenhui Yang, Yong Li, Chris Rizos

IEEE Transactions on Aerospace and Electronic Systems · 2026 · 62 · 9301-9314

DOI: 10.1109/taes.2026.3680890 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Aerospace and Electronic Systems, 2026. Verfasst von Wenhui Yang und 2 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Autonomous vehicles rely on the integration of the inertial navigation system (INS) and the global navigation satellite system (GNSS) for most transportation applications. However, GNSS signals are often degraded in challenging environments such as urban canyons and tunnels, etc. To achieve seamless navigation with continuity and safety, it is necessary to deploy an auxiliary or backup navigation system that does not rely on GNSS. This article proposes a dual microelectromechanical INS (DMI) architecture comprising both rotating and fixed inertial measurement units (IMUs), and utilizes vehicle chassis information to aid the INS during GNSS outages. In terms of multisensor fusion, traditional optimal multisensor fusion takes the centralized Kalman filter (CKF) as a global optimal criterion. The decentralized Kalman filter, such as the well-known federated Kalman filter (FKF), aims to achieve the same accuracy as a CKF, while reducing the computational load. This article innovatively applies an improved decentralized filtering method—Global optimal filter (GOF), and develops a GOF-based DMI integrated solution. This solution optimizes state estimation by fully utilizing multisource data through spatial projection and orthogonalization. The field test results have demonstrated that the GOF-based DMI integrated solution has significant advantages in most evaluation metrics. Specifically, it achieves accuracy improvements of 29.1% in position, 30% in velocity, 22.2% in attitude, and 58.3% in heading over the CKF-based DMI integrated solution. Compared to the FKF-based DMI integrated solution, the accuracy improvements are even more substantial: 70.4% in position, 36.3% in velocity, 70.5% in horizontal attitude, and 72.5% in heading. In addition, the introduction of extended karlsruhe institute of technology and toyota technological institute at chicago (KITTI) evaluation metrics also indicates that the GOF-based DMI integration solution has significant advantages compared to the other two solutions.

Zitieren

APA

Yang, W., Li, Y., & Rizos, C. (2026). GOF-Based DMI Integrated Solution to Seamless Navigation for Autonomous Vehicles During GNSS Outages. IEEE Transactions on Aerospace and Electronic Systems, 62, 9301-9314. https://doi.org/10.1109/taes.2026.3680890

Vancouver

Yang W, Li Y, Rizos C. GOF-Based DMI Integrated Solution to Seamless Navigation for Autonomous Vehicles During GNSS Outages. IEEE Transactions on Aerospace and Electronic Systems. 2026;62:9301-9314. doi:10.1109/taes.2026.3680890

BibTeX
@article{Yang2026,
  author = {Yang, Wenhui and Li, Yong and Rizos, Chris},
  title = {GOF-Based DMI Integrated Solution to Seamless Navigation for Autonomous Vehicles During GNSS Outages},
  journal = {IEEE Transactions on Aerospace and Electronic Systems},
  year = {2026},
  volume = {62},
  pages = {9301--9314},
  doi = {10.1109/taes.2026.3680890}
}
Kostenlos starten →