Challenging data sets for point cloud registration algorithms

Abstract

International audienceThe number of registration solutions in the literature has bloomed recently. The iterative closest point, for example, could be considered as the backbone of many laser-based localization and mapping systems. Although they are widely used, it is a common challenge to compare registration solutions on a fair base. The main limitation is to overcome the lack of accurate ground truth in current data sets, which usually cover environments only over a small range of organization levels. In computer vision, the Stanford 3D Scanning Repository pushed forward point cloud registration algorithms and object modeling fields by providing high-quality scanned objects with precise localization. We aim to provide similar high-caliber working material to the robotic and computer vision communities but with sceneries instead of objects. We propose eight point cloud sequences acquired in locations covering the environment diversity that modern robots are susceptible to encounter, ranging from inside an apartment to a woodland area. The core of the data sets consists of 3D laser point clouds for which supporting data (Gravity, Magnetic North and GPS) are given for each pose. A special effort has been made to ensure global positioning of the scanner within mm-range precision, independent of environmental conditions. This will allow for the development of improved registration algorithms when mapping challenging environments, such as those found in real-world situations

An Improved RANSAC for 3D Point Cloud Plane Segmentation Based on Normal Distribution Transformation Cells

“…The scene itself is of a simple room with some furniture inside. The other two datasets captured by an MLS device were obtained from [41]. In these two datasets, the main sensor of the MLS device was a laser rangefinder (Hokuyo UTM-30 LX) mounted on a tilting unit.…”

A Review of Point Cloud Registration Algorithms for Mobile Robotics

“…But instead of comparing full algorithms with several differences, we can assess the merit of each independent variation. In past works, we have proposed a methodology for this assessment based on real-world datasets [Pomerleau et al, 2012b]. Those works attempt to consolidate the advantages and disadvantages of different algorithms.…”

Global Registration of 3D LiDAR Point Clouds Based on Scene Features: Application to Structured Environments

“…The first dataset is named "apartment" and is available online on the ASL (Autonomous System Lab) website [38]. It has been acquired by the Swiss Federal Institute of Technology in Zürich to test the robustness of algorithms when exploring a dynamic environment.…”