Real-time 6-DOF multi-session visual SLAM over large-scale environments

McDonald, John; Kaess, Michael; Cadena, César; Neira, José; Leonard, John J.

doi:10.1016/j.robot.2012.08.008

Cited by 81 publications

(54 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…"Weak links" are removed when place recognition is able to close loops, making it prone to errors when the place recognition closes loops incorrectly. Similarly, McDonald et al [15] presented a multi-session 6 DOF Visual SLAM system using "anchor nodes". In their approach place recognition is assumed to be perfect and its output is trusted every time.…”

Section: Related Workmentioning

confidence: 99%

Robust loop closing over time for pose graph SLAM

Latif

Cadena

Neira

2013

The International Journal of Robotics Research

197

120

View full text Add to dashboard Cite

Long term autonomous mobile robot operation requires considering place recognition decisions with great caution. A single incorrect decision that is not detected and reconsidered can corrupt the environment model that the robot is trying to build and maintain. This work describes a consensus-based approach to robust place recognition over time, that takes into account all the available information to detect and remove past incorrect loop closures. The main novelties of our work are: (1) the ability of realizing that, in light of new evidence, an incorrect past loop closing decision has been made; the incorrect information can be removed thus recovering the correct estimation with a novel algorithm; (2) extending our proposal to incremental operation; and (3) handling multi-session, spatially related or unrelated scenarios in a unified manner. We demonstrate our proposal, the RRR algorithm, on different odometry systems, e.g. visual or laser using different front-end loop closing techniques. For our experiments we use the efficient graph optimization framework g2o as back-end. We back our claims with several experiments carried out on real data, in single and multi-session experiments showing better results than those obtained by state of the art methods, comparisons against whom are also presented.

show abstract

Section: Related Workmentioning

confidence: 99%

Robust loop closing over time for pose graph SLAM

Latif

Cadena

Neira

2013

The International Journal of Robotics Research

197

120

View full text Add to dashboard Cite

show abstract

“…Different features and robust landmarks extracted from 3D images as points of interest and as references for image mapping and scan registrations have commonly been used for different multi-sessional SLAM (Simultaneous Localization And Mapping) algorithms [30]. This approach works well in simple repetitive paths.…”

Section: D Scan Registrationmentioning

confidence: 99%

Automatic Removal of Imperfections and Change Detection for Accurate 3D Urban Cartography by Classification and Incremental Updating

2013

View full text Add to dashboard Cite

In this article, we present a new method of automatic 3D urban cartography in which different imperfections are progressively removed by incremental updating, exploiting the concept of multiple passages, using specialized functions. In the proposed method, the 3D point clouds are first classified into three main object classes: permanently static, temporarily static and mobile, using a new point matching technique. The temporarily static and mobile objects are then removed from the 3D point clouds, leaving behind a perforated 3D point cloud of the urban scene. These perforated 3D point clouds obtained from successive passages (in the same place) on different days and at different times are then matched together to complete the 3D urban landscape. The changes occurring in the urban landscape over this period of time are detected and analyzed using cognitive functions of similarity, and the resulting 3D cartography is progressively modified accordingly. The specialized functions introduced help to remove the different imperfections, due to occlusions, misclassifications and different changes occurring in the environment over time, thus ncreasing the robustness of the method. The results, evaluated on real data, demonstrate that not only is the resulting 3D cartography accurate, containing only the exact permanent features free from imperfections, but the method is also suitable for handling large urban scenes.Remote Sens. 2013, 5 3702

show abstract

“…Different features and robust landmarks extracted from 3D images as points of interest and as references for image mapping [14] and scan registrations have commonly been used for different multi-sessional SLAM (Simultaneous Localization And Mapping) algorithms [15]. This approach works well in simple repetitive paths.…”

Section: D Scan Registrationmentioning

confidence: 99%

Automatic change detection and incremental updating for accurate 3D urban cartography

Aijazi

Checchin

Trassoudaine

2013

16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013)

View full text Add to dashboard Cite

In this paper, we present a new method of automatic change detection for 3D urban cartography which then progressively incorporates these changes by incremental updating, exploiting the concept of multiple passages. In the proposed method the 3D point clouds, obtained in each passage, are first classified into 2 main object classes: Permanent and Temporary using a voxel-segmentation based classification method and inference based on basic characteristics. The Temporary objects are removed from the 3D point clouds which are then merged together to leave behind a unified perforated 3D point cloud of the urban scene. These perforated 3D point clouds obtained from different passages (in the same place) at different days and times are matched together to complete the 3D urban landscape by incremental updating. Different man-made or natural changes occurring in the urban landscape over this period of time are detected and analyzed using cognitive functions of similarity and the resulting 3D cartography is progressively modified accordingly. The results, evaluated on real data, not only demonstrate the efficacy of the change detection method but also show that these changes are effectively incorporated ensuring that the resulting 3D cartography is updated and contains only the exact permanent features. It is also shown that the proposed method is easily applicable and well suited for handling large urban scenes.

show abstract

Real-time 6-DOF multi-session visual SLAM over large-scale environments

Cited by 81 publications

References 30 publications

Robust loop closing over time for pose graph SLAM

Robust loop closing over time for pose graph SLAM

Automatic Removal of Imperfections and Change Detection for Accurate 3D Urban Cartography by Classification and Incremental Updating

Automatic change detection and incremental updating for accurate 3D urban cartography

Contact Info

Product

Resources

About