GLIO2

GLIO2: A GPU-Parallelized Tightly-Coupled LiDAR–Inertial–GNSS System for Robust and Real-Time Global Localization and Mapping

📢 News

• 📄 2026-10-08: Paper released on arXiv.

📄 Related Papers

• GLIO2: A GPU-Parallelized Tightly-Coupled LiDAR–Inertial–GNSS System for Robust and Real-Time Global Localization and Mapping
• GLIO: Tightly-Coupled GNSS/LiDAR/IMU Integration for Continuous and Drift-free State Estimation of Intelligent Vehicles in Urban Areas

1. Introduction

GLIO2 is a GPU-parallelized, tightly-coupled LiDAR–inertial–GNSS system for robust and real-time global localization and mapping in challenging environments such as long bridges and urban canyons.

1.1 Demo video

🌉 Cross-Sea Bridge LiDAR degeneracy · high-speed driving	🏙️ Urban Canyon Severe GNSS multipath
🚁 Facade-Cleaning UAV Autonomous cleaning on a 110 m high-rise facade	🐕 Robotic Guide Dog Globally referenced navigation in pedestrian scenes

1.2 Applications

Autonomous Facade-Cleaning UAV

Robotic Guide-Dog Navigation

2. Code Release

The code will be open-sourced once the paper is accepted.

3. Acknowledgements

GLIO2 builds on these excellent open-source projects: GLIO, GLIM, GTSAM, gnss_comm, and RTKLIB. We also thank the authors of the UrbanNav dataset.

4. Contact

For further inquiries or assistance, please contact qiiiii.zhang@connect.polyu.hk.