文献类型：期刊文献

英文题名：Multi-LiDAR-Inertial SLAM with Temporally-Coherent Online Calibration

作者：Wang, Liansheng[1]; Zhang, Xinke[1]; Ye, Hangbo[1]; Wang, Chaojie[1]; Pan, Yihan[1]; Yi, Jianjun[1]

机构：[1] dept. of Mechanical Engineering, East China University of Science and Technology, ShangHai, China

年份：2025

期号：2025

起止页码：186

外文期刊名：International Conference on Intelligent Robotics and Control Engineering, IRCE

收录：EI(收录号：20260920152633)

语种：英文

外文关键词：Computational efficiency - Continuous time systems - Mapping - Optical radar - Robotics - Uncertainty analysis

摘要：Recent advancements in sensor fusion have expanded the applications of multi-LiDAR systems in localization and mapping. However, integrating heterogeneous sensor observations from diverse modalities and perspectives presents significant challenges for simultaneous localization and mapping (SLAM) systems. This paper proposes a novel continuous-time multi-LiDAR-inertial SLAM framework with adaptive online calibration that addresses three critical challenges: 1) A pointwise uncertainty modeling approach is developed to handle asynchronous measurements from unsynchronized LiDARs with varying sampling patterns; 2) A continuous-time formulation ensures spatial consistency across multiple LiDARs while enabling distributed real-time online extrinsic calibration to accommodate dynamic sensor configuration changes; 3) An automatic consistency detection criterion is established to intelligently trigger recalibration procedures. The proposed system synergistically integrates filter-based and graph optimization-based methodologies within a unified framework, achieving computational efficiency without compromising accuracy. Extensive evaluations on custom-collected datasets demonstrate that our system completes automatic extrinsic calibration within 10 seconds, with data collection accounting for 95% of the process duration while the calibration itself completes in negligible time. This work advances the state-of-the-art in multi-sensor SLAM by providing robust calibration maintenance and consistent mapping capabilities in complex operational environments. ?2025 IEEE.