PALoc: Advancing SLAM Benchmarking With Prior-Assisted 6-DoF Trajectory Generation and Uncertainty Estimation
- 26 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 29 (6) , 4297-4308
Abstract
Accurately generating ground truth (GT) trajectories is essential for simultaneous localization and mapping (SLAM) evaluation, particularly under varying environmental conditions. This study presents PALoc, a systematic approach that leverages a prior map-assisted framework for the first-time generation of dense six-degree-of-freedom GT poses, significantly enhancing the fidelity of SLAM benchmarks across both indoor and outdoor environments. Our method excels in handling degenerate and stationary conditions frequently encountered in SLAM datasets, thereby increasing robustness and precision. A critical feature of PALoc is the detailed derivation of covariance within the factor graph, enabling an in-depth analysis of pose uncertainty propagation. This analysis plays a pivotal role in illustrating specific pose uncertainty and in elevating trajectory reliability from both theoretical and practical perspectives. In addition, we provide an open-source toolbox for the criteria of map evaluation, facilitating the indirect assessment of overall trajectory precision. Experimental results show at least a 30% improvement in map accuracy and a 20% increase in direct trajectory accuracy compared to the iterative closest point algorithm across diverse environments, with substantially enhanced robustness Our publicly available solution, PALoc, extensively applied in the FusionPortable dataset, is geared toward SLAM benchmark augmentation and represents a significant advancement in SLAM evaluation.Keywords
All Related Versions
Funding Information
- Basic and Applied Basic Research Foundation of Guangdong Province (2021B1515120032)
- The Hong Kong University of Science and Technology (Guangzhou) Joint Funding Program (2024A03J0618)
- Project of Hetao Shenzhen–Hong Kong Science and Technology Innovation Cooperation Zone (HZQB-KCZYB-2020083)
This publication has 46 references indexed in Scilit:
- R3LIVE: A Robust, Real-time, RGB-colored, LiDAR-Inertial-Visual tightly-coupled state Estimation and mapping package2022
- Greedy-Based Feature Selection for Efficient LiDAR SLAM2021
- The Newer College Dataset: Handheld LiDAR, Inertial and Vision with Ground Truth2020
- LINS: A Lidar-Inertial State Estimator for Robust and Efficient Navigation2020
- Monocular Direct Sparse Localization in a Prior 3D Surfel Map2020
- Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping2020
- Are We Ready for Autonomous Drone Racing? The UZH-FPV Drone Racing Dataset2019
- Tightly Coupled 3D Lidar Inertial Odometry and Mapping2019
- Large-Scale Data for Multiple-View StereopsisInternational Journal of Computer Vision, 2016
- Scan registration for autonomous mining vehicles using 3D‐NDTJournal of Field Robotics, 2007