Zuo, Runheng (Shanghai Jiao Tong University), Li, Zexing (Shanghai JiaoTong University), Wang, Yafei (Shanghai Jiao Tong University), Sun, Shi (Shanghai Jiao Tong University), Li, Ruoyao (Shanghai Jiao Tong University), Wu, Mingyu (Shanghai Jiao Tong University)

OSM-Pose: Leveraging OpenStreetMap for LiDAR Based Global Pose Estimation in Urban Autonomous Driving Scenarios

Scheduled for presentation during the Video Session "On-Demand Video Presentations" (VP-VP), Saturday, November 22, 2025, 08:00−18:00, On-Demand Platform

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on April 2, 2026

Abstract

Global LiDAR based localization remains a challenging problem in environments where global navigation satellite systems (GNSS) are unavailable and no pre-built highdefinition maps are accessible. Therefore, we explore the use of OpenStreetMap (OSM), a public available and crowd-sourced map resource, as a lightweight and cost-effective alternative for LiDAR based localization. We propose a two-stage LiDAROSM fusion framework to enable global localization. In the first stage, local odometry estimates are projected into the Universal Transverse Mercator (UTM) coordinate by aligning them with prior road network information through a least squares fitting approach. Initial global trajectories are then generated by associating each pose with the nearest OSM road node. To mitigate observation inconsistency, the second stage refines the pose estimation using a Rotation Prioritized Iterative Closest Point (RPICP) mechanism. RPICP decouples rotation from translation, leveraging building boundary features to achieve more accurate registration between real-time LiDAR scans and prior OSM. The proposed method is validated on KITTI dataset and achieves 1.69 m 2D mean absolute trajectory error in Sequence 08, showing competence among localization methods based on LiDAR-OSM association.