Integrity-First Navigation and Safety-Quantifiable FGO

This program develops positioning methods that estimate both state and trust. We study non-Gaussian error modeling, fault detection, protection-level computation, and integrity constraints inside factor graph optimization for GNSS, GNSS/INS, and multi-sensor localization.

Why it matters. Safety-critical autonomy needs navigation systems that can say not only “Where am I?” but also “Is this answer safe to use?” Without an integrity layer, a confident but wrong position can be more dangerous than no position at all.

Funding & recognition.

• RGC Research Impact Fund R5009-21 — HK$4.5M, PI, 2021–2026 (the only RIF held as PI in the lab’s history)
• 2023 ION Per Enge Early Achievement Award — international single-awardee per year
• US12123961B2 patent (granted 2024) — 3D LiDAR Aided GNSS NLOS Detection and Correction (Hsu, Wen)

Applications. Autonomous driving, aviation, UAVs and low-altitude systems, urban service robotics.

Representative publications. Integrity-Constrained Factor Graph Optimization (Xia, Wen, Hsu, NAVIGATION 2024); Principal Gaussian Overbound for Heavy-tailed Error Bounding (Yan, Zhong, Hsu, IEEE TAES 2024); Fault Detection Algorithm for Gaussian Mixture Noises: An Application in Lidar/IMU Integrated Localization (Yan et al., NAVIGATION 2025).