Odometry Estimation from Velodyne LiDAR Point Clouds

This software implements a novel method of odometry estimation from Velodyne LiDAR point cloud scans, i.e. the change in pose and orientation of the scanning device over time. The implemented method overcomes the most painful issues of Velodyne data - the sparsity and the quantity of data points - in an efficient way, enabling more precise registration. Alignment of two consequent Velodyne scans which yields the final odometry is based on a random sampling of the clouds using Collar Line Segments (CLS) and identification of correspondences between these line segments. Using the matched line segments, the software is able to estimate a precise transformation aligning the two Velodyne scans.