Hello

This is a forked repository of M-LOAM.

In this we add extra script and configuration files to quickly run the experiments on NTU VIRAL dataset.

Instructions

1. After you have cloned and compiled the package, please cd to the launch folder by running roscd mloam/launch.

2. Modify the script runall_ntuviral.sh to declare the output directory via the variable EPOC_DIR, and the path to the dataset via the DATASET_LOCATION variable. Note that the bag files should be orgnanized like in the figure below:

3. Run bash runall_ntuviral.sh to automatically launch the algo with the bag files one by one.

4. The log files will be created in .csv format and you can use matlab or python to calculate the localization error against the ground truth on these logs. An example of the analysis can be found in the M-LOAM/analysis/ntuviral_mloam folder. Just run the MATLAB script checkall_parallel.m and the results will be printed out.

If you use the NTU VIRAL dataset in your work, please cite our work using the BibTex code at the dataset's website. Please also cite M-LOAM's publication below.

M-LOAM

Robust Odometry and Mapping for Multi-LiDAR Systems with Online Extrinsic Calibration

M-LOAM is a robust system for multi-LiDAR extrinsic calibration, real-time odometry, and mapping. Without manual intervention, our system can start with several extrinsic-uncalibrated LiDARs, automatically calibrate their extrinsics, and provide accurate poses as well as a globally consistent map.

Authors: Jianhao Jiao, Haoyang Ye, Yilong Zhu, Linxin Jiang, Ming Liu from RAM-LAB, HKUST

Project website: https://ram-lab.com/file/site/m-loam

Videos:

(Video link for mainland China friends: Video)

Related Papers in Solving Different Subproblems for Multi-LiDAR Systems

• Robust Odometry and Mapping for Multi-LiDAR Systems with Online Extrinsic Calibration, Jianhao Jiao, Haoyang Ye, Yilong Zhu, Ming Liu, IEEE Transactions on Robotics (T-RO), 2021. pdf

  • Tackle the extrinsic calibration, multi-LiDAR fusion, pose drift, and mapping uncertainty.

• Greedy-Based Feature Selection for Efficient LiDAR SLAM, Jianhao Jiao, Yilong Zhu, Haoyang Ye, Huaiyang Huang, Peng Yun, Linxin Jiang, Lujia Wang, Ming Liu, International Conference on Robotics and Automation (ICRA) 2021 , Xi An, China. pdf

  • Tackle the algorithm latency issue.

• MLOD: Awareness of Extrinsic Perturbation in Multi-LiDAR 3D Object Detection for Autonomous Driving, Jianhao Jiao*, Peng Yun*, Lei Tai, Ming Liu, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020. pdf

  • Tackle the multi-LiDAR-based 3D object detection against the hardware failure (injected large extrinsic perturbation).

If you use M-LOAM for your academic research, please cite one of our paper. bib

1. Prerequisites

1.1 Ubuntu and ROS

Ubuntu 64-bit 16.04 or 18.04. ROS Kinetic or Melodic. ROS Installation

1.2. Ceres Solver && Eigen3 && GLOG

 ./setup/install_eigen3_ceres.sh

1.3. OpenMP

  sudo apt install libomp-dev

1.4. PCL-1.8

2. Build M-LOAM on ROS

  mkdir -p ~/catkin_ws/src
  cd ~/catkin_ws/src
  git clone https://github.com/gogojjh/M-LOAM.git
  catkin build mloam
  source ~/catkin_ws/devel/setup.bash

3. Example

• Datasets collected with different platforms:

  1. Simulation Robot (SR)
  2. Real Handheld Device (RHD)
  3. Real Vechile (RV)
  4. Oxford RoboCar (OR)

• Run M-LOAM and baseline methods

  1. We provide a script to perform batch testing of M-LOAM with baseline methods
  2. Enter the script folder: roscd mloam/script/
  3. Modify the python script: run_mloam.py for specific platforms with correct path
  4. Modify the shell files for methods in xx_main.sh
  5. Run the python script:
  • python2 run_mloam.py -program=single_test -sequence=SR -start_idx=0 -end_idx=4
    • You will broadcast the SR01.bag, SR02.bag, SR03.bag, SR04.bag, SR05.bag respectively to test the M-LOAM system.
  • python2 run_mloam.py -program=single_test -sequence=RHD -start_idx=0 -end_idx=2
    • You will broadcast the RHD02lab.bag, RHD03garden.bag, RHD04building.bag respectively to test the M-LOAM system.
  • python2 run_mloam.py -program=single_test -sequence=RHD -start_idx=1 -end_idx=1
    • You will broadcast the RV01.bag to test the M-LOAM system.

4. System pipeline

This could help you to understand the pipeline of M-LOAM (loop closure part is not finished).

And you can also refer to M-LOAM's pipeline for a more detailed diagram and code review.

5. Acknowledgements

Thanks for these great works from which we learned to develop M-LOAM

• LOAM (J. Zhang and S. Singh. LOAM: Lidar Odometry and Mapping in Real-time) and its advanced version: A-LOAM;
• LEGO-LOAM
• LIO-MAPPING
• VINS-MONO
• Lidar Perception Library

Thanks for Ming Cao for providing a clear diagram and code review of M-LOAM

6. Licence

The source code is released under GPLv3 license.

For any technical issues, please contact Dr. Jianhao Jiao [email protected]. For commercial inquiries, please contact Prof.Ming Liu [email protected].