Distributed SOLiD SLAM

📰 News

• We release a modified version that changed LIO-SAM to LIORF in here!!

   $ git clone --branch liorf https://github.com/sparolab/Distributed-SOLiD-SLAM.git
  

📂 What is Distributed SOLiD SLAM?

• Distributed SOLiD SLAM is a Distributed SOLiD-based LiDAR SLAM Framework, which is a modified version of LIO-SAM and DiSCo-SLAM. (Scan Context → SOLiD)
• The information exchange between robots is made through ROS-based communication. (More detailed in here!!)
• SOLiD, which is a lightweight descriptor enables fast communication between robots.

📦 Dependencies

• Ubuntu 20.04
• GTSAM (Develop version)
• libnabo 1.0.7

📦 How to use the Distributed SOLiD SLAM?

Linux

$ cd ~/catkin_ws/src
$ git clone https://github.com/sparolab/Distributed-SOLiD-SLAM.git
$ cd ..
$ catkin_make
$ source devel/setup.bash
$ roslaunch lio_sam run.launch
$ rosbag play (your dataset).bag

Docker

$ cd ~/catkin_ws/src
$ git clone https://github.com/sparolab/Distributed-SOLiD-SLAM.git
$ docker pull cokr6901/distributed_solid_slam
$ docker run --privileged --gpus all \
-it --name distributed_solid_slam --ipc=host --shm-size=512M \
--device=/dev/video0:/dev/video0 -v /tmp/.X11-unix:/tmp/.X11-unix:ro \
-e DISPLAY=unix$DISPLAY -v /root/.Xauthority:/root/.Xauthority \
--env="QT_X11_NO_MITSHM=1" \
-v ~/catkin_ws/src/:/home/test_ws/src -v (your dataset folder path)/:/home/test_ws/storage cokr6901/distributed_solid_slam:latest
$ cd /home/test_ws/
$ catkin_make
$ source devel/setup.bash
$ roslaunch lio_sam run.launch
$ rosbag play (your dataset).bag

⚙️ Parameters

Extrinsic (LiDAR -> IMU)

extrinsicTrans: [0.0, 0.0, 0.0]
extrinsicRot: [-1, 0, 0,
	  0, 1, 0,
	  0, 0, -1]
extrinsicRPY: [0,  1, 0,
	 -1, 0, 0,
	  0, 0, 1]
		  

SOLiD

mapfusion:
    solid:
	knn_feature_dim: 40
	max_range: 80
	num_sector: 60
	num_height: 64
	num_nearest_matches: 50
	num_match_candidates: 1
	fov_up: 2.0
	fov_down: -24.8

⚙️ Utils

Generate a multi-robot rosbag from a single-robot rosbag using a Python script. (reference: here )

$ python3 split.bag -i (input.bag) -o (output.bag)

You can edit these lines.

topics = ['/points_raw', '/imu_raw', '/gps/fix']       # Rostopic names
split_places = [90, 180, 290]		               # 0(start)-90-180-290(final)
robot_names =  ['/jackal0', '/jackal1', '/jackal2']    # Robot names (jackal0:0-90 / jackal1:90-180 / jackal2:180-290)

⚙️ ETC

• You can see the results of the Park dataset (i.e. DiSCo SLAM dataset) here!!
  • You should modify parameters to fit the Robot and LiDAR!!
    • Extrinsic (For Robot)
    • SOLiD (For VLP-16)

       mapfusion:
           solid:
       	num_height: 16
       	fov_up: 15.0
       	fov_down: -15.0
      

💡 To DO

• Save the transformed paths.
• Change the LiDAR-Odometry. (LIO-SAM → FAST-LIO2)
• Add the outlier rejection. (e.g. RANSAC)
• Add the traversability mapping.

📒 Citation

  @article{kim2024narrowing,
    title={Narrowing your FOV with SOLiD: Spatially Organized and Lightweight Global Descriptor for FOV-constrained LiDAR Place Recognition},
    author={Kim, Hogyun and Choi, Jiwon and Sim, Taehu and Kim, Giseop and Cho, Younggun},
    journal={IEEE Robotics and Automation Letters},
    year={2024},
    publisher={IEEE}
  }

📧 Contact

• Hogyun Kim (hg.kim@inha.edu)
• Juwon Kim (lambertkim@naver.com)

👏 Special Thanks

• We appreciate Prof. Brendan Englot's RobustFieldAutonomyLab, particularly Yewei Huang, for publishing the DiSCo-SLAM.