LiDAR-Based Object-Level SLAM for Autonomous Vehicles

Cao, Bingyi; Mendoza, Ricardo; Philipp, Andreas; Göhring, Daniel

doi:10.1109/iros51168.2021.9636299

Cited by 11 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We evaluate our global data association method using candidate loop closure pairs from KITTI [20] sequences 00, 02, 05, and 08. We compare our method, called GraffMatch, with a pole-only method [13] based on 2D cluster matching [37], and a plane-only method [34] that attempts to match planes via nearest neighbor search on CP parameterization [16] followed by RANSAC [30]. We adapt the pole-only method [13] to 3D and denote it as PoleMatch, while the plane-only method is denoted PlaneMatch.…”

Section: Methodsmentioning

confidence: 99%

“…We compare our method, called GraffMatch, with a pole-only method [13] based on 2D cluster matching [37], and a plane-only method [34] that attempts to match planes via nearest neighbor search on CP parameterization [16] followed by RANSAC [30]. We adapt the pole-only method [13] to 3D and denote it as PoleMatch, while the plane-only method is denoted PlaneMatch. The algorithms are implemented in MATLAB 1 and executed on an i9-7920X CPU with 64 GB RAM.…”

Section: Methodsmentioning

confidence: 99%

“…Schlichting and Brenner [36] extend this descriptor to include upright planes, but effectively treat poles and planes as 2D points and lines. Cao et al [13] perform object-level SLAM using poles, walls, and parked cars as landmarks and propose to use pole positions within a scan to create a scan signature used for detecting loops. Upon detecting a pair of scans as a loop candidate, clusters of pole-points are matched [37] and a rigid transformation is estimated in a RANSAC framework.…”

Section: Related Workmentioning

confidence: 99%

“…second. To maintain the benefits of geometric data, but to reduce the storage and computational costs of large point maps, some lidar odometry and SLAM systems use geometric primitives like lines and planes instead of points [9]- [13].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Global Data Association for SLAM with 3D Grassmannian Manifold Objects

C.¹,

How²

2022

Preprint

View full text Add to dashboard Cite

Using pole and plane objects in lidar SLAM can increase accuracy and decrease map storage requirements compared to commonly-used point cloud maps. However, place recognition and geometric verification using these landmarks is challenging due to the requirement for global matching without an initial guess. Existing works typically only leverage either pole or plane landmarks, limiting application to a restricted set of environments. We present a global data association method for loop closure in lidar scans using 3D line and plane objects simultaneously and in a unified manner. The main novelty of this paper is in the representation of line and plane objects extracted from lidar scans on the manifold of affine subspaces, known as the affine Grassmannian. Line and plane correspondences are matched using our graph-based data association framework and subsequently registered in the leastsquares sense. Compared to pole-only approaches and planeonly approaches, our 3D affine Grassmannian method yields a 71 % and 325 % increase respectively to loop closure recall at 100 % precision on the KITTI dataset and can provide frame alignment with less than 10 cm and 1 deg of error.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Global Data Association for SLAM with 3D Grassmannian Manifold Objects

C.¹,

How²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Additionally, individual object models, often desirable for downstream tasks such as asset mapping, are not used in the SLAM optimization. By contrast, other works build object-level maps using pre-collected 3D models [11], or basic shapes such as points [1], cuboids [4], ellipsoids [2], cylinders [7], a set of semantic keypoints [12], or a combination of 2D shapes [13]. Some efforts also account for ambiguity in data association [1], and the usage of semantic information in identifying moving objects [14], generating hierarchical descriptors for loop closure [15], and minimizing odometry drifts to assist long-range navigation [16].…”

Section: A Metric-semantic Mappingmentioning

confidence: 99%

Active Metric-Semantic Mapping by Multiple Aerial Robots

Liu¹,

Ankit²,

Cladera³

et al. 2022

Preprint

View full text Add to dashboard Cite

Traditional approaches for active mapping focus on building geometric maps. For most real-world applications, however, actionable information is related to semantically meaningful objects in the environment. We propose an approach to the active metric-semantic mapping problem that enables multiple heterogeneous robots to collaboratively build a map of the environment. The robots actively explore to minimize the uncertainties in both semantic (object classification) and geometric (object modeling) information. We represent the environment using informative but sparse object models, each consisting of a basic shape and a semantic class label, and characterize uncertainties empirically using a large amount of real-world data. Given a prior map, we use this model to select actions for each robot to minimize uncertainties. The performance of our algorithm is demonstrated through multirobot experiments in diverse real-world environments. The proposed framework is applicable to a wide range of realworld problems, such as precision agriculture, infrastructure inspection, and asset mapping in factories.

show abstract