radiation_simulation

ROS 2 and Gazebo-based radiation simulation and radiation-aware navigation package.

Overview

This repository provides a ROS 2 package for simulating radiation-aware mobile robot operation in Gazebo.
It includes custom Gazebo plugins for radiation sources, radiation sensors, and attenuating obstacles, along with radiation map generation, visualization, local/global radiation-aware costmap layers, and waypoint-based autonomous navigation.

The radiation-related Gazebo plugin implementation in this repository was developed with reference to prior work on simulating ionising radiation in Gazebo for robotic nuclear inspection.
The layered costmap design for radiation-aware navigation was also informed by prior work on real-time radiation avoidance using layered costmaps for mobile robots.

This repository is organized as a single ROS 2 package at the repository root.

Features

• Gazebo-based radiation source, sensor, and obstacle plugins
• Radiation map generation and visualization
• Radiation-aware local/global costmap layers for Nav2
• Waypoint-based autonomous navigation
• Preconfigured experiment worlds for reactor_room and single_source
• Radiation logging and post-processing utilities for cumulative exposure analysis

Dependencies

• ROS 2 Humble
• Python 3.10 / C++17
• Gazebo Classic / gazebo_ros
• Nav2
• RViz2
• TurtleBot3 description
• pluginlib
• tf2 / tf2_ros / tf2_geometry_msgs
• OpenCV

Recommended installation:

sudo apt update
sudo apt install -y \
  ros-humble-gazebo-ros-pkgs \
  ros-humble-nav2-bringup \
  ros-humble-nav2-msgs \
  ros-humble-nav2-map-server \
  ros-humble-nav2-lifecycle-manager \
  ros-humble-rviz2 \
  ros-humble-turtlebot3-description \
  ros-humble-pluginlib \
  ros-humble-tf2-ros \
  ros-humble-tf2-geometry-msgs \
  ros-humble-robot-state-publisher \
  libgazebo11-dev \
  qtbase5-dev \
  libopencv-dev

Additional External Packages

If you want to run frontier exploration experiments based on m-explore-ros2 / explore_lite, install that package separately in your ROS 2 workspace instead of expecting it to be included in this repository.

Installation

mkdir -p ~/<workspace_name>/src
cd ~/<workspace_name>/src
git clone https://github.com/RCILab/RCI_radiation.git

cd ~/<workspace_name>
colcon build --symlink-install
source install/setup.bash

Usage

1) Reactor room simulation

Launch the Gazebo simulation, robot model, radiation map generation, and visualization in the reactor_room world:

ros2 launch radiation_simulation reactor_room_spawn.launch.py

Run waypoint-based navigation in the reactor_room environment:

ros2 launch radiation_simulation reactor_room_waypoint.launch.py

2) Single-source simulation

Launch the simplified single_source world:

ros2 launch radiation_simulation single_source_spawn.launch.py

Run waypoint-based navigation and radiation logging in the single_source world:

ros2 launch radiation_simulation single_source_waypoint.launch.py

3) Optional additional launch files

The package also includes additional launch files such as:

ros2 launch radiation_simulation explore_radiation.launch.py
ros2 launch radiation_simulation map_sweeper.launch.py

Experiment Worlds

reactor_room

reactor_room is a more complex environment for radiation simulation and navigation experiments. The reactor_room world assets included in this repository are based on source code obtained from:

Simulating Ionising Radiation in Gazebo for Robotic Nuclear Inspection Challenges

single_source

single_source is a simplified rectangular-room world with a radiation source placed at the center of the room. It is useful for more intuitive and visually interpretable demonstrations of radiation-aware navigation and avoidance behavior.

Logging and Post-processing

Radiation logging

single_source_waypoint.launch.py includes scripts/radiation_logger.py, which records radiation intensity and robot pose during navigation.

The logger stores CSV files under:

data/cumulation/<world_name>/

For the current single_source waypoint setup, the logged CSV is configured as:

data/cumulation/single_source/avoid_thres30_60deg_ver4.csv

The logger records:

• time
• radiation intensity
• robot position
• robot yaw

Post-processing

You can analyze cumulative exposure results using:

python3 scripts/cumulation_analyzer.py

The analyzer is configured for the single_source experiment and compares:

• no_avoid_vel10.csv
• avoid_thres30_60deg_ver4.csv

It produces three plots:

1. Time-Intensity
2. Trajectory
3. Time-Cumulative Exposure

Example:

Repository Structure

RCI_radiation/
├── .gitignore
├── CMakeLists.txt
├── LICENSE
├── README.md
├── package.xml
├── plugin_description.xml
├── config/
├── data/
├── include/
│   └── radiation_simulation/
├── launch/
├── map/
├── models/
├── rviz/
├── scripts/
├── src/
├── urdf/
└── worlds/

Main Components

• Radiation Source Plugin Publishes source intensity from Gazebo models and supports runtime updates.

• Radiation Sensor Plugin Detects radiation intensity using source distance, attenuation, and obstacle interaction, and publishes measurement topics.

• Radiation Obstacle Plugin Publishes attenuation coefficients for obstacle models based on material properties.

• Radiation Map Generator Builds and publishes radiation-related map and visualization data.

• Radiation Visualization Node Displays total and source-wise detected radiation information.

• Radiation Local / Global Costmap Layers Extends Nav2 layered costmaps with radiation-aware cost information.

• Waypoint Navigation Executes waypoint-based autonomous navigation using Nav2.

Notes

• This README intentionally focuses on simulation, plugins, radiation-aware costmaps, navigation, and logging utilities.
• Source localization-related components are not covered here.
• This repository itself is the radiation_simulation ROS 2 package.
• Some logs, messages, comments, or auxiliary outputs in this repository are written in Korean.

References

The radiation-related plugin design was developed with reference to:

• Wright, T., West, A., Licata, M., Hawes, N., and Lennox, B. Simulating Ionising Radiation in Gazebo for Robotic Nuclear Inspection Challenges. Robotics, 2021, 10(3):86. DOI: 10.3390/robotics10030086

The radiation-aware navigation and layered costmap design was developed with reference to:

• West, A., Wright, T., Tsitsimpelis, I., Groves, K., Joyce, M. J., and Lennox, B. Real-Time Avoidance of Ionising Radiation Using Layered Costmaps for Mobile Robots. Frontiers in Robotics and AI, 2022, 9:862067. DOI: 10.3389/frobt.2022.862067

Software Registration

The autonomous navigation simulation program for radiation map construction developed in this project has been officially registered for copyright.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

Maintainer: jth8090 (jth8090@khu.ac.kr)
Lab: RCI Lab @ Kyung Hee University