[ICRA 2025 Best Paper Award Finalist]

Wooseong Yang · Hyesu Jang · Ayoung Kim

• [12/05/2025]: Full source code is uploaded.
• [24/04/2025]: Our paper is selected as the finalist for ICRA 2025 Best Paper Award!
• [29/01/2025]: Our paper is accepted to ICRA 2025.

We recommend using the provided Docker environment (Ubuntu 20.04) for testing our code.

1. Clone our repository

   mkdir -p gorio_ws/src && cd gorio_ws/src
   git clone https://github.com/wooseongY/Go-RIO .

2. Build the Docker image

   cd docker
   sudo chmod +x build.sh
   ./build.sh

3. Modify the DATA_DIR in the run.sh as the appropriate data folder
4. Run the Docker container and build the package

   sudo chmod +x run.sh
   ./run.sh
   catkin_make && source devel/setup.bash

5. Modify the bag file path in the rosbag_play_<sequence>.launch file as "/root/data/<your_bag_directory>", which contains your proper bag file.
6. Launch our algorithm and enjoy :)

   roslaunch gorio <launch file name>.launch
   rostopic pub /command std_msgs/String "output_aftmapped"

If you use our paper for any academic work, please cite our paper.

@INPROCEEDINGS {wsyang-2025-icra,
    author={Wooseong Yang and Hyesu Jang and Ayoung Kim},
    title={Ground-Optimized 4D Radar-Inertial Odometry via Continuous Velocity Integration using Gaussian Process},
    booktitle={Proceedings of the IEEE International Conference on Robotics and Automation (ICRA)},
    year={2025},
    month={May.},
    address={Atlanta},
}

If you have any questions, please contact:

• Wooseong Yang (wseongy15@gmail.com)

This work was supported by the Robotics and AI (RAI) Institute, and in part by the MOITE, Korea (1415187329,20024355).

And thanks for authors of UGPM, 4DRadarSLAM, Patchwork and REVE.