HaLeGO-LOAM

HaLeGO-LOAM: A Real-Time LiDAR SLAM System that uses Homogenized Point Cloud Sampling and Adaptive Feature Extraction

• HaLeGO-LOAM is an optimized LiDAR-based SLAM system that enhances feature extraction and point representation, eliminating the need for high-precision IMU sensors.

Embedded Real-time Lidar SLAM:

Sc-LeGO-LOAM (22 Sensors) + ALeGO-LOAM (2020 IEEE Intelligent Vehicles Symposium (IV))

• SC-LeGO-LOAM is a lightweight, user-friendly, and fast LiDAR SLAM system, allowing for efficient loop detection and scan context creation with just two API functions.
• ALeGO-LOAM leverages pole maps and fast point cloud feature extraction to significantly enhance vehicle localization accuracy and trajectory correction in self-driving systems.

Features

• Homogenized Point Cloud Sampling: Ensures clearer edge and planar extraction, boosting path estimation accuracy.
• Adaptive Feature Extraction: Maintains point cloud density at vast distances and minimizes short-ranges feature sampling.

Performance

• Consistently outperforms LeGO-LOAM, LiTAMIN2, and MULLS-SLAM on the KITTI dataset, offering significantly improved accuracy. In LGSVL simulations, it demonstrates 7-15% better edge position accuracy.

Application

• Achieving approximately 30 fps on NVIDIA Xavier, HaLeGO-LOAM stands as a viable solution for embedded applications.

• Current version: September, 2023.

Dependencies

• All dependencies are same as LeGO-LOAM (i.e., ROS, PCL, and GTSAM).
• C++14.

How to use

• Place the directory HaLeGO-LOAM under user catkin work space
• For example,

  cd ~/catkin_ws/src
  git clone https://github.com/louis960126/HaLeGO-LOAM.git
  cd ..
  catkin_make
  source devel/setup.bash
  roslaunch lego_loam run.launch
  

Comparison

KITTI sequence 00 Blue: HaLeGO-LOAM Red: Ground Truth

Dataset	HaLeGO-LOAM	LeGO-LOAM
KITTI sequence 00	0.47%	2.17%

Demo

• video (https://www.youtube.com/watch?v=UxahkPNjOMg&ab_channel=Sheng-weiLee)
• Dataset (Ouster 128)(Ouster 32) (Uploading)

Cite ALeGO-LOAM

@INPROCEEDINGS{9304747,
  author={Lee, Sheng-Wei and Lin, Peng-Wei and Fu, Yuan-Ting and Hsu, Chih-Ming and Chan, Chen-Yu and Lin, Jhih-Hong and Chiang, Yen-Hung},
  booktitle={2020 IEEE Intelligent Vehicles Symposium (IV)}, 
  title={Improving vehicle localization using pole-like landmarks extracted from 3-D lidar scans}, 
  year={2020},
  pages={2052-2057},
  organization={IEEE}
}

Cite SC-LeGO-LOAM

@INPROCEEDINGS { gkim-2018-iros,
  author = {Kim, Giseop and Kim, Ayoung},
  title = { Scan Context: Egocentric Spatial Descriptor for Place Recognition within {3D} Point Cloud Map },
  booktitle = { Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems },
  year = { 2018 },
  month = { Oct. },
  address = { Madrid }
}

and

@inproceedings{legoloam2018,
  title={LeGO-LOAM: Lightweight and Ground-Optimized Lidar Odometry and Mapping on Variable Terrain},
  author={Shan, Tixiao and Englot, Brendan},
  booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages={4758-4765},
  year={2018},
  organization={IEEE}
}

Contact

• Maintainer: ([email protected])

NOTE

• Code is currently being organized and will be continuously uploaded.