A failure detection method for 3D LiDAR based localization

Huan Yin; Li Tang; Xiaqing Ding; Yue Wang; Rong Xiong

doi:10.1109/CAC48633.2019.8996777

A failure detection method for 3D LiDAR based localization

Huan Yin, Li Tang, Xiaqing Ding, Yue Wang, Rong Xiong

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

9 Citations (Scopus)

Abstract

3D LiDAR based localization is a very common method for robot self-positioning on prior maps. Currently, few research works str focus on the failure detection of laser localization, which is a critical part in autonomous systems for mobile robots or intelligent vehicles. In this paper, we propose a statistical learning based method for localization failure detection, formulating the problem as a binary classification task. Specifically, we first extract features to measure the align quality of point clouds, which describe the geometric properties in local and global levels. Then a logistic regression model is trained and can detect the localization failures in new environments. We employ our self-collected dataset to demonstrate the effectiveness of the proposed method. The results show that the logistic regression model can achieve high accuracy on failure detection for 3D LiDAR based localization.

Original language	English
Title of host publication	Proceedings - 2019 Chinese Automation Congress, CAC 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	4559-4563
Number of pages	5
ISBN (Electronic)	9781728140940
DOIs	https://doi.org/10.1109/CAC48633.2019.8996777
Publication status	Published - Nov 2019
Externally published	Yes
Event	2019 Chinese Automation Congress, CAC 2019 - Hangzhou, China Duration: 22 Nov 2019 → 24 Nov 2019

Publication series

Name	Proceedings - 2019 Chinese Automation Congress, CAC 2019

Conference

Conference	2019 Chinese Automation Congress, CAC 2019
Country/Territory	China
City	Hangzhou
Period	22/11/19 → 24/11/19

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Keywords

3D LiDAR
failure detection
localization
logistic regression
mobile robot

Access to Document

10.1109/CAC48633.2019.8996777

Cite this

Yin, H., Tang, L., Ding, X., Wang, Y., & Xiong, R. (2019). A failure detection method for 3D LiDAR based localization. In Proceedings - 2019 Chinese Automation Congress, CAC 2019 (pp. 4559-4563). Article 8996777 (Proceedings - 2019 Chinese Automation Congress, CAC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CAC48633.2019.8996777

@inproceedings{9a9883978399426d8da2817d5016a95b,

title = "A failure detection method for 3D LiDAR based localization",

abstract = "3D LiDAR based localization is a very common method for robot self-positioning on prior maps. Currently, few research works str focus on the failure detection of laser localization, which is a critical part in autonomous systems for mobile robots or intelligent vehicles. In this paper, we propose a statistical learning based method for localization failure detection, formulating the problem as a binary classification task. Specifically, we first extract features to measure the align quality of point clouds, which describe the geometric properties in local and global levels. Then a logistic regression model is trained and can detect the localization failures in new environments. We employ our self-collected dataset to demonstrate the effectiveness of the proposed method. The results show that the logistic regression model can achieve high accuracy on failure detection for 3D LiDAR based localization.",

keywords = "3D LiDAR, failure detection, localization, logistic regression, mobile robot",

author = "Huan Yin and Li Tang and Xiaqing Ding and Yue Wang and Rong Xiong",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 Chinese Automation Congress, CAC 2019 ; Conference date: 22-11-2019 Through 24-11-2019",

year = "2019",

month = nov,

doi = "10.1109/CAC48633.2019.8996777",

language = "English",

series = "Proceedings - 2019 Chinese Automation Congress, CAC 2019",

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

pages = "4559--4563",

booktitle = "Proceedings - 2019 Chinese Automation Congress, CAC 2019",

}

Yin, H, Tang, L, Ding, X, Wang, Y & Xiong, R 2019, A failure detection method for 3D LiDAR based localization. in Proceedings - 2019 Chinese Automation Congress, CAC 2019., 8996777, Proceedings - 2019 Chinese Automation Congress, CAC 2019, Institute of Electrical and Electronics Engineers Inc., pp. 4559-4563, 2019 Chinese Automation Congress, CAC 2019, Hangzhou, China, 22/11/19. https://doi.org/10.1109/CAC48633.2019.8996777

A failure detection method for 3D LiDAR based localization. / Yin, Huan; Tang, Li; Ding, Xiaqing et al.
Proceedings - 2019 Chinese Automation Congress, CAC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 4559-4563 8996777 (Proceedings - 2019 Chinese Automation Congress, CAC 2019).

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

TY - GEN

T1 - A failure detection method for 3D LiDAR based localization

AU - Yin, Huan

AU - Tang, Li

AU - Ding, Xiaqing

AU - Wang, Yue

AU - Xiong, Rong

PY - 2019/11

Y1 - 2019/11

N2 - 3D LiDAR based localization is a very common method for robot self-positioning on prior maps. Currently, few research works str focus on the failure detection of laser localization, which is a critical part in autonomous systems for mobile robots or intelligent vehicles. In this paper, we propose a statistical learning based method for localization failure detection, formulating the problem as a binary classification task. Specifically, we first extract features to measure the align quality of point clouds, which describe the geometric properties in local and global levels. Then a logistic regression model is trained and can detect the localization failures in new environments. We employ our self-collected dataset to demonstrate the effectiveness of the proposed method. The results show that the logistic regression model can achieve high accuracy on failure detection for 3D LiDAR based localization.

AB - 3D LiDAR based localization is a very common method for robot self-positioning on prior maps. Currently, few research works str focus on the failure detection of laser localization, which is a critical part in autonomous systems for mobile robots or intelligent vehicles. In this paper, we propose a statistical learning based method for localization failure detection, formulating the problem as a binary classification task. Specifically, we first extract features to measure the align quality of point clouds, which describe the geometric properties in local and global levels. Then a logistic regression model is trained and can detect the localization failures in new environments. We employ our self-collected dataset to demonstrate the effectiveness of the proposed method. The results show that the logistic regression model can achieve high accuracy on failure detection for 3D LiDAR based localization.

KW - 3D LiDAR

KW - failure detection

KW - localization

KW - logistic regression

KW - mobile robot

UR - https://www.webofscience.com/wos/woscc/full-record/WOS:000679040104116

UR - https://openalex.org/W3005754415

UR - https://www.scopus.com/pages/publications/85080034740

U2 - 10.1109/CAC48633.2019.8996777

DO - 10.1109/CAC48633.2019.8996777

M3 - Conference Paper published in a book

T3 - Proceedings - 2019 Chinese Automation Congress, CAC 2019

SP - 4559

EP - 4563

BT - Proceedings - 2019 Chinese Automation Congress, CAC 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 Chinese Automation Congress, CAC 2019

Y2 - 22 November 2019 through 24 November 2019

ER -

A failure detection method for 3D LiDAR based localization

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this