Abstract
Although visual SLAM (simultaneous localization and mapping) methods obtain very accurate results using optimization of residual errors defined with respect to the matching features, the SLAM systems based on 3-D laser (LiDAR) data commonly employ variants of the iterative closest points algorithm and raw point clouds as the map representation. However, it is possible to extract from point clouds features that are more spatially extended and more meaningful than points: line segments and/or planar patches. In particular, such features provide a natural way to represent human-made environments, such as urban and mixed indoor/outdoor scenes. In this paper, we perform an analysis of the advantages of a LiDAR-based SLAM that employs high-level geometric features in large-scale urban environments. We present a new approach to the LiDAR SLAM that uses planar patches and line segments for map representation and employs factor graph optimization typical to state-of-the-art visual SLAM for the final map and trajectory optimization. The new map structure and matching of features make it possible to implement in our system an efficient loop closure method, which exploits learned descriptors for place recognition and factor graph for optimization. With these improvements, the overall software structure is based on the proven LOAM concept to ensure real-time operation. A series of experiments were performed to compare the proposed solution to the open-source LOAM, considering different approaches to loop closure computation. The results are compared using standard metrics of trajectory accuracy, focusing on the final quality of the estimated trajectory and the consistency of the environment map. With some well-discussed reservations, our results demonstrate the gains due to using the high-level features in the full-optimization approach in the large-scale LiDAR SLAM.
Highlights
The knowledge of the robot pose and the availability of an environment model are enabling conditions for efficient task and motion planning, reasoning about the environment’s semantics, and human–machine interaction in autonomous robot vehicles
We propose new features, a new map structure, and its optimization procedure, the overall software architecture of PlaneLOAM retains a similarity to the LiDAR odometry and mapping (LOAM) system [9], which combines real-time scan-to-scan pose tracking and slower, but more accurate scan-to-map localization
Since it is not possible to review here the overwhelming amount of work in simultaneous localization and mapping (SLAM) published in the last two decades, we focus on the prior works that are directly related to our research in three aspects: the general architecture and map representation, the methods used for registration of consecutive scans, and the approaches to detect and close the loops whenever the robot revisits an already mapped area
Summary
The knowledge of the robot pose and the availability of an environment model are enabling conditions for efficient task and motion planning, reasoning about the environment’s semantics, and human–machine interaction in autonomous robot vehicles. It is possible to efficiently extract from the LiDAR point clouds some higher-level geometric features, namely line segments and planar patches. In order to exploit re-observations of the already visited parts of the environment, we implemented a loop detection mechanism, which is based on SegMap [8] This approach learns robust descriptors of the segmented point clouds, which are used to determine the re-observed locations in the map. The proposed solution is evaluated on three different datasets employing different LiDAR sensors We compared it against the open-source LOAM system and tested three versions of PlaneLOAM 2.0: without loop closures, following the LOAM approach but with higher-level features, and with loop closures implemented either using a simple pose graph or employing the factor graph approach.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
