Indoor loop closure detection based on geometric features

Wensong Jiang; Zhiyuan Zhu; Zai Luo; Li Yang

doi:10.1117/12.2611713

19 November 2021 Indoor loop closure detection based on geometric features

Wensong Jiang, Zhiyuan Zhu, Zai Luo, Li Yang

Proceedings Volume 12059, Tenth International Symposium on Precision Mechanical Measurements; 120590L (2021) https://doi.org/10.1117/12.2611713
Event: Tenth International Symposium on Precision Mechanical Measurements, 2021, Qingdao, China

Abstract

Loop closure detection (LCD) is an important and challenging task in simultaneous localization and mapping (SLAM). To improve the efficiency and accuracy of an indoor LiDAR LCD, a geometric feature-based method is proposed. First, the rotation invariance of the keyframe pair to be detected is realized by the Fourier transform. Second, the main features of the keyframe are extracted and fitted into geometric features, which are stored in a ring-shaped semantic image. Third, the similarity score of the two keyframes is obtained by calculating the distance between two corresponding ring semantic images, which is verified by the ICP algorithm. The experimental result shows that the time cost of each detection is reduced to less than 30ms without sacrificing the detection precision by using our suggested method.

Citation Download Citation

Wensong Jiang, Zhiyuan Zhu, Zai Luo, and Li Yang "Indoor loop closure detection based on geometric features", Proc. SPIE 12059, Tenth International Symposium on Precision Mechanical Measurements, 120590L (19 November 2021); https://doi.org/10.1117/12.2611713

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available