Rui Zhang, Rob Worley, Sarah Edwards, Jonathan Aitken, Sean R. Anderson, and Lyudmila Mihaylova
This work introduces a novel approach to visual Simultaneous Localization and Mapping (vSLAM) that incorporates Cylindrical Information for localisation and mapping in sewer pipe networks. Our method builds upon the robust foundation provided by ORB-SLAM2, extending its capabilities to exploit the unique environmental features of sewer pipe networks for improved accuracy and efficiency.
Please refer to our publication for a detailed explanation of our approach and its advantages:
R. Zhang, R. Worley, S. Edwards, J. Aitken, S. R. Anderson, and L. Mihaylova, "Visual Simultaneous Localization and Mapping for Sewer Pipe Networks Leveraging Cylindrical Regularity," in IEEE Robotics and Automation Letters, vol. 8, no. 6, pp. 3406-3413, June 2023, doi: 10.1109/LRA.2023.3268013. PDF.
If you utilize CRORB (Monocular) in your academic research, please cite our work as follows:
@ARTICLE{10103583,
author={Zhang, Rui and Worley, Rob and Edwards, Sarah and Aitken, Jonathan and Anderson, Sean R. and Mihaylova, Lyudmila},
journal={IEEE Robotics and Automation Letters},
title={Visual Simultaneous Localization and Mapping for Sewer Pipe Networks Leveraging Cylindrical Regularity},
year={2023},
volume={8},
number={6},
pages={3406-3413},
doi={10.1109/LRA.2023.3268013}
}
The setup and implementation details for CRORB remain consistent with those established for ORB-SLAM2. For comprehensive instructions on prerequisites, building the project, and running monocular examples, please refer to the original ORB-SLAM2.