Real-time tightly coupled GNSS and IMU integration via Factor Graph Optimization

A team of researchers has published a new paper on arXiv detailing a breakthrough in real-time navigation for autonomous systems in dense cities. The work, "Real-time tightly coupled GNSS and IMU integration via Factor Graph Optimization," addresses the critical challenge of reliable positioning where GPS (GNSS) signals are frequently blocked by buildings, suffer from multipath reflections, or have rapidly changing satellite geometry. While offline Factor Graph Optimization (FGO) methods for fusing GPS and inertial measurement unit (IMU) data are known for their accuracy, this new formulation makes the process causal and real-time, a key requirement for moving vehicles and robots.

The technical core of the method is an incremental optimization approach with fixed-lag marginalization, which allows the system to process sensor data continuously and in real-time, rather than in post-processing batches. This enables causal state estimation—meaning the system's understanding of its position is based only on past and present data, not future information. The team rigorously evaluated their algorithm's performance using the UrbanNav dataset, a benchmark known for simulating highly urbanized, GNSS-degraded environments. This advancement paves the way for more reliable and accurate autonomous navigation for drones, delivery robots, and self-driving cars in complex urban canyons where traditional GPS alone fails.