Accurate and Robust Object SLAM with 3D Quadric Landmark Reconstruction in Outdoors

Rui Tian; Yunzhou Zhang; Yonghui Feng; Linghao Yang; Zhenzhong Cao; Sonya Coleman; Dermot Kerr

doi:10.1109/LRA.2021.3137896

Accurate and Robust Object SLAM with 3D Quadric Landmark Reconstruction in Outdoors

Rui Tian, Yunzhou Zhang, Yonghui Feng, Linghao Yang, Zhenzhong Cao, Sonya Coleman, Dermot Kerr

Research output: Contribution to journal › Article › peer-review

18 Citations (Scopus)

376 Downloads (Pure)

Abstract

Object-oriented SLAM is a popular technology in autonomous driving and robotics. In this letter, we propose a stereo visual SLAM with a robust quadric landmark representation method.The system consists of four components, including deep learning detection, quadric landmark initialization, object data association and object pose optimization. State-of-the-art quadric-based SLAM algorithms always face observation-related problems and are sensitive to observation noise, which limits their application in outdoor scenes. To solve this problem, we propose a quadric initialization method based on the separation of the quadric parameters method, which improves the robustness to observation noise. The sufficient object data association algorithm and object-oriented optimization with multiple cues enable a highly accurate object pose estimation that is robust to local observations. Experimental results show that the proposed system is more robust to observation noise and significantly outperforms current state-of-the-art methods in outdoor environments. In addition, the proposed system demonstrates real-time performance.

Original language	English
Pages (from-to)	1534-1541
Number of pages	8
Journal	IEEE Robotics and Automation Letters
Volume	7
Issue number	2
Early online date	23 Dec 2021
DOIs	https://doi.org/10.1109/LRA.2021.3137896
Publication status	Published (in print/issue) - 30 Apr 2022

Bibliographical note

Funding Information:
This work was supported in part by the National Natural Science Foundation of China under Grants 61973066 and 61471110, in part by the Major Science and Technology Projects of Liaoning Province under Grant 2021JH1/10400049, in part by the Fundation of Key Laboratory of Aerospace System Simulation under Grant 6142002200301, in part by the Fundation of Key Laboratory of Equipment Reliability under Grant WD2C20205500306, and in part by the Major Science and Technology Innovation Engineering Projects of Shandong Province under Grant 2019JZZY010128.

Publisher Copyright:
© 2016 IEEE.

Keywords

Cameras
Ellipsoids
Localization
Optimization
Semantics
Simultaneous localization and mapping
SLAM
Three-dimensional displays
Vision-Based Navigation
Visualization
Vision-based navigation

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/LRA.2021.3137896

Quadric_object_slam_resubmit_UlsterAuthor Accepted Manuscript, 2.27 MB

Cite this

@article{29583a4d8c6046698f39910e3cc14ede,

title = "Accurate and Robust Object SLAM with 3D Quadric Landmark Reconstruction in Outdoors",

abstract = "Object-oriented SLAM is a popular technology in autonomous driving and robotics. In this letter, we propose a stereo visual SLAM with a robust quadric landmark representation method.The system consists of four components, including deep learning detection, quadric landmark initialization, object data association and object pose optimization. State-of-the-art quadric-based SLAM algorithms always face observation-related problems and are sensitive to observation noise, which limits their application in outdoor scenes. To solve this problem, we propose a quadric initialization method based on the separation of the quadric parameters method, which improves the robustness to observation noise. The sufficient object data association algorithm and object-oriented optimization with multiple cues enable a highly accurate object pose estimation that is robust to local observations. Experimental results show that the proposed system is more robust to observation noise and significantly outperforms current state-of-the-art methods in outdoor environments. In addition, the proposed system demonstrates real-time performance.",

keywords = "Cameras, Ellipsoids, Localization, Optimization, Semantics, Simultaneous localization and mapping, SLAM, Three-dimensional displays, Vision-Based Navigation, Visualization, Vision-based navigation",

author = "Rui Tian and Yunzhou Zhang and Yonghui Feng and Linghao Yang and Zhenzhong Cao and Sonya Coleman and Dermot Kerr",

note = "Funding Information: This work was supported in part by the National Natural Science Foundation of China under Grants 61973066 and 61471110, in part by the Major Science and Technology Projects of Liaoning Province under Grant 2021JH1/10400049, in part by the Fundation of Key Laboratory of Aerospace System Simulation under Grant 6142002200301, in part by the Fundation of Key Laboratory of Equipment Reliability under Grant WD2C20205500306, and in part by the Major Science and Technology Innovation Engineering Projects of Shandong Province under Grant 2019JZZY010128. Publisher Copyright: {\textcopyright} 2016 IEEE.",

year = "2022",

month = apr,

day = "30",

doi = "10.1109/LRA.2021.3137896",

language = "English",

volume = "7",

pages = "1534--1541",

journal = "IEEE Robotics and Automation Letters",

issn = "2377-3766",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - Accurate and Robust Object SLAM with 3D Quadric Landmark Reconstruction in Outdoors

AU - Tian, Rui

AU - Zhang, Yunzhou

AU - Feng, Yonghui

AU - Yang, Linghao

AU - Cao, Zhenzhong

AU - Coleman, Sonya

AU - Kerr, Dermot

N1 - Funding Information: This work was supported in part by the National Natural Science Foundation of China under Grants 61973066 and 61471110, in part by the Major Science and Technology Projects of Liaoning Province under Grant 2021JH1/10400049, in part by the Fundation of Key Laboratory of Aerospace System Simulation under Grant 6142002200301, in part by the Fundation of Key Laboratory of Equipment Reliability under Grant WD2C20205500306, and in part by the Major Science and Technology Innovation Engineering Projects of Shandong Province under Grant 2019JZZY010128. Publisher Copyright: © 2016 IEEE.

PY - 2022/4/30

Y1 - 2022/4/30

N2 - Object-oriented SLAM is a popular technology in autonomous driving and robotics. In this letter, we propose a stereo visual SLAM with a robust quadric landmark representation method.The system consists of four components, including deep learning detection, quadric landmark initialization, object data association and object pose optimization. State-of-the-art quadric-based SLAM algorithms always face observation-related problems and are sensitive to observation noise, which limits their application in outdoor scenes. To solve this problem, we propose a quadric initialization method based on the separation of the quadric parameters method, which improves the robustness to observation noise. The sufficient object data association algorithm and object-oriented optimization with multiple cues enable a highly accurate object pose estimation that is robust to local observations. Experimental results show that the proposed system is more robust to observation noise and significantly outperforms current state-of-the-art methods in outdoor environments. In addition, the proposed system demonstrates real-time performance.

AB - Object-oriented SLAM is a popular technology in autonomous driving and robotics. In this letter, we propose a stereo visual SLAM with a robust quadric landmark representation method.The system consists of four components, including deep learning detection, quadric landmark initialization, object data association and object pose optimization. State-of-the-art quadric-based SLAM algorithms always face observation-related problems and are sensitive to observation noise, which limits their application in outdoor scenes. To solve this problem, we propose a quadric initialization method based on the separation of the quadric parameters method, which improves the robustness to observation noise. The sufficient object data association algorithm and object-oriented optimization with multiple cues enable a highly accurate object pose estimation that is robust to local observations. Experimental results show that the proposed system is more robust to observation noise and significantly outperforms current state-of-the-art methods in outdoor environments. In addition, the proposed system demonstrates real-time performance.

KW - Cameras

KW - Ellipsoids

KW - Localization

KW - Optimization

KW - Semantics

KW - Simultaneous localization and mapping

KW - SLAM

KW - Three-dimensional displays

KW - Vision-Based Navigation

KW - Visualization

KW - Vision-based navigation

UR - http://www.scopus.com/inward/record.url?scp=85122073507&partnerID=8YFLogxK

U2 - 10.1109/LRA.2021.3137896

DO - 10.1109/LRA.2021.3137896

M3 - Article

AN - SCOPUS:85122073507

SN - 2377-3766

VL - 7

SP - 1534

EP - 1541

JO - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

IS - 2

ER -

Accurate and Robust Object SLAM with 3D Quadric Landmark Reconstruction in Outdoors

Abstract

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this