DaLiTI

[RA-L 2025] Degradation-aware LiDAR-Thermal-Inertial SLAM

This repository contains code for a Degradation aware LiDAR-Thermal-Inertial SLAM, which is accepted by IEEE RA-L, June 2025.

1. Dependency

• ROS noetic (tested)
• Basic libraries for common SLAM (e.g. openCV 4, pcl-1.10, ceres-solver-1.14 etc.)
• gtsam 4.0.3

2. Using Docker (Recommended)

2.1 Build Docker Image

First, navigate to your workspace and build the Docker image:

cd YOUR_NAME_OF_WORKSPACE
docker build -t daliti:latest -f src/DaLiTI/Dockerfile .

2.2 Run Docker Container

Use the provided startup script to launch the container with RViz support:

cd src/DaLiTI
./start_docker.sh

The script will:

• Mount the source code directory (src/DaLiTI) into the container
• Enable X11 forwarding for RViz visualization
• Start an interactive zsh shell with ROS environment pre-configured

2.3 Run Examples in Docker

Once inside the container, you can run the demo:

roslaunch daliti demo.launch rosbag_file_with_path:="/ws/YOUR/PATH/TO/ROSBAG/XXX.bag"

Note: Make sure to place your rosbag files in a location accessible to the container, or mount additional directories when starting the container.

3. Run Examples (Local Installation)

mkdir -p YOUR_NAME_OF_WORKSPACE/src
cd YOUR_NAME_OF_WORKSPACE/src
git clone https://github.com/HITSZ-NRSL/DaLiTI.git
cd ..

catkin_make or catkin build

roslaunch daliti demo.launch rosbag_file_with_path:="YOUR/PATH/TO/ROSBAG/XXX.bag"

For validation, the hitsz degration dataset is provided: dowload link

4. Acknowledgements

DaLiTI is partly modified codes form LVI-SAM, FAST-LIO2, VINS-Mono, ESKF_LIO and rebvo. Thanks for the contributers.

Contributer: Yufeng Liu

5. Citation

If you use our work, please cite:

@ARTICLE{11045071,
         author={Wang, Yu and Liu, Yufeng and Chen, Lingxu and Chen, Haoyao and Zhang, Shiwu},
         journal={IEEE Robotics and Automation Letters},
         title={Degradation-Aware LiDAR-Thermal-Inertial SLAM},
         year={2025},
         volume={10},
         number={8},
         pages={8035-8042},
         doi={10.1109/LRA.2025.3581127}}

LICENSE

The source code is released under GPLv3 license.