A Plane-based Approach for Indoor Point Clouds Registration

Favre, Ketty; Pressigout, Muriel; Marchand, Éric; Morin, Luce

doi:10.1109/icpr48806.2021.9412379

Cited by 18 publications

(8 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further recent examples of laser-based SLAM approaches making use of the existence of planes include [23][24][25][26]. Similar registration procedures to ours are [27], which uses plane-to-plane correspondences for pre-registration and point-to-plane correspondences afterward, and [28], which uses point-to-point, as well as plane-to-plane correspondences based on their availability. Two more recent and very interesting SLAM approaches which specialize more on the massivley produced LiVOX devices, are "Loam-livox" [3] and "Livox-mapping" [29].…”

Section: Related Workmentioning

confidence: 96%

Unconventional Trajectories for Mobile 3D Scanning and Mapping

Arzberger¹,

Zevering²,

Bredenbeck³

et al. 2023

Autonomous Mobile Mapping Robots

View full text Add to dashboard Cite

State-of-the-art LiDAR-based 3D scanning and mapping systems focus on scenarios where good sensing coverage is ensured, such as drones, wheeled robots, cars, or backpack-mounted systems. However, in some scenarios more unconventional sensor trajectories come naturally, e.g., rolling, descending, or oscillating back and forth, but the literature on these is relatively sparse. As a result, most implementations developed in the past are not able to solve the SLAM problem in such conditions. In this chapter, we propose a robust offline-batch SLAM system that is able to address more challenging trajectories, which are characterized by weak angles of incidence and limited FOV while scanning. The proposed SLAM system is an upgraded version of our previous work and takes as input the raw points and prior pose estimates, yet the latter are subject to large amounts of drift. Our approach is a two-staged algorithm where in the first stage coarse alignment is fast achieved by matching planar polygons. In the second stage, we utilize a graph-based SLAM algorithm for further refinement. We evaluate the mapping accuracy of the algorithm on our own recorded datasets using high-resolution ground truth maps, which are available from a TLS.

show abstract

Section: Related Workmentioning

confidence: 96%

Unconventional Trajectories for Mobile 3D Scanning and Mapping

Arzberger¹,

Zevering²,

Bredenbeck³

et al. 2023

Autonomous Mobile Mapping Robots

View full text Add to dashboard Cite

show abstract

“…Their approach generates maps consisting of polygonal planar patches and is shown to work well on a variety of indoor and outdoor environments, outperforming both Generalized ICP and LOAM, while running at rates between 2 and 10 Hz (depending on the number of points vs. planes used). Favre et al [14] also present an approach for plane-based point cloud registration which is shown to outperform traditional point-based methods while running at 1-3 Hz on various datasets, but does not consider the problem of SLAM and map generation. Additionally, a variety of approaches for generating plane-based maps from dense depth information exist [8], [15], [16].…”

Section: Related Workmentioning

confidence: 99%

PlaneSLAM: Plane-based LiDAR SLAM for Motion Planning in Structured 3D Environments

Dai¹,

Lund²,

Gao³

2022

Preprint

View full text Add to dashboard Cite

LiDAR sensors are a powerful tool for robot simultaneous localization and mapping (SLAM) in unknown environments, but the raw point clouds they produce are dense, computationally expensive to store, and unsuited for direct use by downstream autonomy tasks, such as motion planning. For integration with motion planning, it is desirable for SLAM pipelines to generate lightweight geometric map representations. Such representations are also particularly wellsuited for man-made environments, which can often be viewed as a so-called "Manhattan world" built on a Cartesian grid. In this work we present a 3D LiDAR SLAM algorithm for Manhattan world environments which extracts planar features from point clouds to achieve lightweight, real-time localization and mapping. Our approach generates plane-based maps which occupy significantly less memory than their point cloud equivalents, and are suited towards fast collision checking for motion planning. By leveraging the Manhattan world assumption, we target extraction of orthogonal planes to generate maps which are more structured and organized than those of existing plane-based LiDAR SLAM approaches. We demonstrate our approach in the high-fidelity AirSim simulator and in realworld experiments with a ground rover equipped with a Velodyne LiDAR. For both cases, we are able to generate high quality maps and trajectory estimates at a rate matching the sensor rate of 10 Hz.

show abstract

“…Airborne LiDAR (Light Detection And Ranging) and photogrammetry point clouds are nowadays employed to produce high-quality features and models. These datasets are very efficient data for different applications in photogrammetry and remote sensing, such as building detection (Yi et al, 2021), 3D building reconstruction (Mahphood and Arefi, 2017, Tarsha Kurdi et al, 2021, Yastikli and Cetin, 2021, digital terrain model (DTM) generation (Mongus and Žalik, 2012), change detection (Liu et al, 2021), ground point filtering (Zeybek and Şanlıoğlu, 2019), point cloud registration (Favre et al, 2021), as well as building boundary extraction (Widyaningrum et al, 2019, Mahphood andArefi, 2022).…”

Section: Introductionmentioning

confidence: 99%

Density-Based Method for Building Detection From Lidar Point Cloud

Mahphood

Arefi

2023

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. In this paper, a new building detection method based on a density of LiDAR point clouds is proposed. In this method, trees, vegetation, and any objects that have points in a vertical plane or column are removed. In the density-based method, a cube is utilized to calculate the density therein. For each point, the cube is used to determine the number of neighbouring points. The density is calculated in two cases: 3D and 2D space. In 3D space, the volumetric density is calculated using the cube. In 2D space, all points are projected onto the horizontal plane, and the surface density is calculated using a square. Next, the two densities are compared and the points with different values in both cases are removed. The method leads to promising results in the removal of vegetation and trees. Moreover, the results achieve more than 94% completeness and correctness at the per-area level.

show abstract

A Plane-based Approach for Indoor Point Clouds Registration

Cited by 18 publications

References 23 publications

Unconventional Trajectories for Mobile 3D Scanning and Mapping

Unconventional Trajectories for Mobile 3D Scanning and Mapping

PlaneSLAM: Plane-based LiDAR SLAM for Motion Planning in Structured 3D Environments

Density-Based Method for Building Detection From Lidar Point Cloud

Contact Info

Product

Resources

About