CoFi: Coarse-to-Fine ICP for LiDAR Localization in an Efficient Long-lasting Point Cloud Map

Abstract

LiDAR odometry and localization has attracted increasing research interest in recent years. In the existing works, iterative closest point (ICP) is widely used since it is precise and efficient. Due to its non-convexity and its local iterative strategy, however, ICP-based method easily falls into local optima, which in turn calls for a precise initialization. In this paper, we propose CoFi, a Coarse-to-Fine ICP algorithm for LiDAR localization. Specifically, the proposed algorithm downsamples the input point sets under multiple voxel resolution, and gradually refines the transformation from the coarse point sets to the fine-grained point sets. In addition, we propose a map based LiDAR localization algorithm that extracts semantic feature points from the LiDAR frames and apply CoFi to estimate the pose on an efficient point cloud map. With the help of the Cylinder3D algorithm for LiDAR scan semantic segmentation, the proposed CoFi localization algorithm demonstrates the state-of-the-art performance on the KITTI odometry benchmark, with significant improvement over the literature.

Mobile Laser Scanning Data Collected under a Forest Canopy with GNSS/INS-Positioned Systems: Possibilities of Processability Improvements

“…More authors [77,[90][91][92] dealt with methods of point cloud misalignment evaluation, partitioning, and registration, including adaptive time-based [77,93] and spacebased [94] partitioning methods. Point cloud registration has also been paid considerable attention [92,94,95], including the coarse-to-fine strategy [90,96,97] and iterative closest point (ICP) algorithm application [92,[98][99][100][101][102]. Clustering points of the misaligned point cloud from the MLS survey of the forest into point clouds with no copies of the scanned objects using time partitioning has previously proved to be efficient [10,21,50].…”