aalto1 untyped-item.component.html
Non-Repetitive Scan Correction for LiDAR Odometry via Spatial Analysis in CERN Facilities
Loading...
Access rights
openAccess
CC BY
CC BY
Creative Commons license
Except where otherwised noted, this item's license is described as openAccess
publishedVersion
URL
Journal Title
Journal ISSN
Volume Title
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Date
Major/Subject
Mcode
Degree programme
Language
en
Pages
17
Series
IEEE Access, Volume 14, pp. 52717-52733
Abstract
Autonomous navigation of mobile robots in large-scale, feature-sparse environments remains a challenging problem for localization and mapping, especially when using non-repetitive scanning LiDAR sensors. Existing LiDAR registration methods, which are largely tailored to repetitive scan patterns, often perform poorly on sparse and irregular point clouds with limited scan-to-scan overlap. To address this limitation, this paper proposes a spatial-context-aware LiDAR registration framework that improves generalized iterative closest point (GICP) performance under non-repetitive scanning conditions using LiDAR-only point clouds, without IMU or wheel odometry integration. The method is validated in multiple CERN facilities, whose long, structurally similar tunnels and large-scale spaces constitute a demanding experimental setting. By extracting robust geometric features, identifying motion patterns, and adaptively combining multiple pose estimation strategies, the proposed framework improves registration robustness and accuracy. Experiments conducted in four distinct environments, including cylindrical tunnels, rectangular laboratories, and dome-shaped chambers, show significant reductions in registration error and improved map quality compared with GICP, wheel odometry, KISS-ICP, and kinematic ICP. Additional evaluation on the Indoor Benchmark of 3D LiDAR SLAM under varying conditions within the same scene further demonstrates competitive performance against LiDAR-IMU fusion methods. Overall, this work offers a practical and effective solution for LiDAR registration and mapping with non-repetitive scans using LiDAR alone. The code and benchmarking dataset are publicly available at: https://github.com/Pejman712/NonRep.git.
Description
Publisher Copyright: © 2013 IEEE.
Other note
Citation
Habibiroudkenar, P, Forkel, D, Ojala, R, Hari Prasanth, S M, Davidsson, D H, Kari, T, Matheson, E & Di Castro, M 2026, 'Non-Repetitive Scan Correction for LiDAR Odometry via Spatial Analysis in CERN Facilities', IEEE Access, vol. 14, pp. 52717-52733. https://doi.org/10.1109/ACCESS.2026.3680632