Narrowing your FOV with SOLiD: Spatially Organized and Lightweight Global Descriptor for FOV-constrained LiDAR Place Recognition

IEEE Robotics and Automation Letters (RA-L)

News!

Solid State LiDAR Place Recognition
(Narrow FOV)

Video

Abstract

We often encounter limited FOV situations due to various factors such as sensor fusion or sensor mount in real-world robot navigation. However, the limited FOV interrupts the generation of descriptions and impacts place recognition adversely. Therefore, we suffer from correcting accumulated drift errors in a consistent map using LiDAR-based place recognition with limited FOV.

Thus, in this paper, we propose a robust LiDAR-based place recognition method for handling narrow FOV scenarios. The proposed method establishes spatial organization based on the range-elevation bin and azimuth-elevation bin to represent places.  In addition, we achieve a robust place description through reweighting based on vertical direction information. Based on these representations, our method enables addressing rotational changes and determining the initial heading.  Additionally, we design a lightweight and fast approach for the onboard autonomy of the robot.

For rigorous validation, the proposed method has been tested across various LiDAR place recognition scenarios (i.e. single session, multi-session, and multi-robot scenarios). 

To the best of our knowledge, we report the first method to cope with the restricted FOV. Our place description and SLAM codes will be released.

Method

 We define the spatial organization and describe SOLiD generation from a 3-D scan and calculation of the distance between the two as represented in the pipeline Figure.

Spatial organization refers to organizing a 3-D space into 2-D bins along the range-elevation and azimuth-elevation directions using polar coordinates and determining representative values for these bins.

Unlike the BEV representation, the spatial organization includes elevation information, allowing for multi-directional analysis of 3-D space.

Therefore, to achieve place recognition within a limited FOV, we utilize spatial organization and create SOLiD by reweighting vertical information.

Code

You can see our code in this Github.

Appendix

You can see the Appendix in this link.

BibTex

@article{kim2024narrowing,

  title={Narrowing your FOV with SOLiD: Spatially Organized and Lightweight Global Descriptor for FOV-constrained LiDAR Place Recognition},

  author={Kim, Hogyun and Choi, Jiwon and Sim, Taehu and Kim, Giseop and Cho, Younggun},

  journal={IEEE Robotics and Automation Letters},

  year={2024},

  publisher={IEEE}

}