Efficient and Accurate Vehicle Localization Based on LiDAR Place Recognition

An efficient and accurate LiDAR place recognition methodology is proposed for vehicle localization. First, the Iris-LOAM to overcome disadvantages of low accuracy loop-closure detection efficiency map construction in existing LOAM-series methods. The method integrates LiDAR-Iris global descriptor Normal Distribution Transform (NDT) registration into module Odometry Mapping (LOAM), thereby improving construction. For shortcomings loading matching efficiency, Random Sample Consensus used remove ground point cloud information. Voxel Grid down sample loaded map. Finally, NDT adopted obtain position Show that has higher than SC-LeGO-LOAM. average time reduced by 39.5%. can be executed achieve