aiqbench analysis
This research introduces a novel multi-submap architecture for lidar odometry that mitigates cumulative registration errors by optimizing constraints across overlapping map segments. The method achieves superior accuracy and consistency on automotive datasets while maintaining real-time performance through retrospective refinement of submaps, marking a significant advancement in autonomous navigation algorithms.
Online Lidar-Only Odometry with Retrospective Refinement of Overlapping Submaps
arXiv:2503.21293v3 Announce Type: replace Abstract: Lidar odometry aims to estimate the ego-motion of a mobile platform from a stream of lidar scans. Traditional scan-to-map approaches register each scan against a single, evolving map, which propagates registration errors over time. To mitigate this, we propose a multi-submap approach where the current scan is registered against multiple overlapping submaps instead of a single static map. By optimizing the resulting constraints in a pose graph, our method enables not only precise estimation of the current pose, but also retrospective refinement of the submaps' anchor points, which improves short-term consistency and long-term accuracy. We demonstrate that our approach achieves significant accuracy gains over several state-of-the art lidar odometry methods on a variety of automotive datasets while simultaneously maintaining real-time performance. Ablation studies confirm the critical role of multiple registrations and retrospective refinement of the map as core factors for our accuracy gains. Code and raw results are available on our public GitHub at https://github.com/Fusion-Goettingen/RAL_2026_Kurda_Online.
Abstract:Lidar odometry aims to estimate the ego-motion of a mobile platform from a stream of lidar scans. Traditional scan-to-map approaches register each scan against a single, evolving map, which propagates registration errors over time. To mitigate this, we propose a multi-submap approach where the current scan is registered against multiple overlapping submaps instead of a single static map. By optimizing the resulting constraints in a pose graph, our method enables not only precise estimation of the current pose, but also retrospective refinement of the submaps' anchor points, which improves short-term consistency and long-term accuracy. We demonstrate that our approach achieves significant accuracy gains over several state-of-the art lidar odometry methods on a variety of automotive datasets while simultaneously maintaining real-time performance. Ablation studies confirm the critical role of multiple registrations and retrospective refinement of the map as core factors for our accuracy gains. Code and raw results are available on our public GitHub at this https URL.
Robotics (cs.RO)
arXiv:2503.21293 [cs.RO]
(or arXiv:2503.21293v3 [cs.RO] for this version)
Full text under the publisher's open licence · 167 words
Related coverage
LV-Calib: LiDAR-Camera Extrinsic Calibration with Boundary-Response Modeling
arXiv · Jun 16, 2026
CU-Multi: A Dataset for Multi-Robot Collaborative Perception
arXiv · Jun 11, 2026
GR-LIO: A Local Ground-Aware LiDAR-Inertial Odometry System Using Body-to-Ground Height
arXiv · Oct 6, 2026
ArborSplat: Online Semantic Gaussian Splatting SLAM for Orchards
arXiv · Sep 24, 2026