TNDF-Fusion: Implicit Truncated Neural Distance Field for LiDAR Dense Mapping and Localization in Large Urban Environments

Chen, Zhiming1; Zhang, Kun1; Chen, Hua2; Wang, Michael Yu3; Zhang, Wei2; Yu, Hongyu4,5

2024-09-01

10.1109/LRA.2024.3427552

IEEE ROBOTICS AND AUTOMATION LETTERS   影响因子和分区

2377-3766

Large-scale 3D mapping is an important task for robotics and autonomous driving. However, mobile robots and autonomous vehicles with limited hardware resources may face issues with large memory consumption. It is challenging to achieve a balance between mapping quality and memory consumption. To address this issue, we propose a new compact implicit neural map representation - the Tri-Pyramid that can infer the Truncated Neural Distance Field (TNDF) given an arbitrary 3D position. Additionally, we introduce a TNDF label rectification method considering both the direction of ground normals and closest surface points to enhance the precision of supervision signals for training with a set of effective loss functions. Experiments on public datasets demonstrated that our method reaches comparable or superior performance for dense mapping while significantly reducing memory consumption compared to previous LiDAR mapping approaches. Furthermore, our study confirms the scalability and adaptability of our approach from room-scale to city-scale scenes. Moreover, we explore the potential of directly leveraging the implicit neural map representation for localization tasks by solving an optimization problem. The experiments showcase the accurate localization capabilities of our method in various scenarios.

SLAM mapping range sensing

SCI ; EI

英语

通讯

Innovation and Technology Commission of HKSAR[ITS/036/21FP] ; Project of Hetao Shenzhen-Hong Kong Science and Technology Innovation Cooperation Zone[HZQB-KCZYB-2020083]

Robotics

Robotics

WOS:001273087700007

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC

Web of Science

1.Hong Kong Univ Sci & Technol, Cheng Kar Shun Robot Inst, Hong Kong, Peoples R China
2.Southern Univ Sci & Technol, Sch Syst Design & Intelligent Mfg, Shenzhen 518055, Peoples R China
3.Great Bay Univ, Sch Engn, Dongguan 523106, Peoples R China
4.Hong Kong Univ Sci & Technol, Robot Inst, Hong Kong 999077, Peoples R China
5.HKUST Shenzhen Hong Kong Collaborat Innovat Res In, Shenzhen, Peoples R China

Chen, Zhiming,Zhang, Kun,Chen, Hua,et al. TNDF-Fusion: Implicit Truncated Neural Distance Field for LiDAR Dense Mapping and Localization in Large Urban Environments[J]. IEEE ROBOTICS AND AUTOMATION LETTERS,2024,9(9).

Chen, Zhiming,Zhang, Kun,Chen, Hua,Wang, Michael Yu,Zhang, Wei,&Yu, Hongyu.(2024).TNDF-Fusion: Implicit Truncated Neural Distance Field for LiDAR Dense Mapping and Localization in Large Urban Environments.IEEE ROBOTICS AND AUTOMATION LETTERS,9(9).

Chen, Zhiming,et al."TNDF-Fusion: Implicit Truncated Neural Distance Field for LiDAR Dense Mapping and Localization in Large Urban Environments".IEEE ROBOTICS AND AUTOMATION LETTERS 9.9(2024).