Authors: Radu-Andrei Cioaca, Cristian Rusu, Paul Irofti, Gianluca Caparra, Andrei-Alexandru Marinache, Florin Stoican
In our latest research, “Real-time loosely coupled GNSS and IMU integration via Factor Graph Optimization,” we address the critical need for reliable Positioning, Navigation, and Timing (PNT) in autonomous systems. While Factor Graph Optimization (FGO) has significantly advanced localization performance, its high computational requirements have traditionally limited its use to offline post-processing.
Our work proposes a loosely coupled architecture that enables FGO-based GNSS and IMU fusion to operate in real-time, even within challenging “urban canyons”. By testing our approach on the UrbanNav-HK-MediumUrban-1 dataset, we demonstrate that this method achieves increased service availability compared to standard batch FGO methods. This paper provides an in-depth analysis of the essential trade-offs between localization accuracy, computational efficiency, and service reliability, offering a practical framework for the next generation of autonomous navigation.
https://arxiv.org/abs/2603.03546
