Project Structure
The repository is organized as follows:
sumoware_bridge/
├── src/ # Source code directory
│ ├── sumoware/ # Main package containing the bridge implementation
│ └── tier4_autoware_msgs/ # Autoware message definitions
├── sumo_docker/ # Docker configuration for SUMO
├── launcher_autoware/ # Modified Autoware launch configurations
├── map/ # Map-related files and utilities
├── example_inputs/ # Example input files
│ ├── map.xodr # Example OpenDRIVE map file
│ └── scenario_params.json # Example scenario parameters
Running the simulation
Tested with WSL (Ubuntu 22.04) on Windows 11 with ROS2 Humble
Clone this repository, recursively
git clone git@github.com:TUM-VT/SumoWare.git --recurse-submodules
cd sumoware
Make sure ROS2 Humble is installed ROS2 Humble
SUMO needs to be installed for map conversion: SUMO Installation
Make sure SUMO_HOME environment variable is set
sudo apt-get install sumo sumo-tools sumo-doc
export SUMO_HOME=/usr/share/sumo
Make sure docker is installed
Make sure rocker is installed
sudo apt-get install python3-rocker
Build sumo docker image with the name "sumo_docker"
docker build ./sumo_docker -t sumo_docker
Source the ROS2 installation
source /opt/ros/humble/setup.bash
Install dependencies
sudo apt install python3-rosdep2
sudo apt install python3-pip
sudo apt-get install ros-humble-tf-transformations
# pip install empy==3.3.4 # uncomment if colcon build fails
pip install requests==2.31.0 # for rocker compatibility
pip install --upgrade transforms3d
pip install commonroad-scenario-designer
rosdep update
rosdep install --from-paths src --ignore-src -r -y
Build the project
sudo apt install python3-colcon-common-extensions
colcon build --symlink-install
Source the project workspace
source install/local_setup.bash
Pull autoware image
docker pull ghcr.io/autowarefoundation/autoware:universe-cuda-20241010-amd64
Run the simulation with example data
ros2 run sumoware scenario_controller --ros-args -p use_sim_time:=true -p scenario_file:="./example_inputs/scenario_params.json" -p map_file:="./example_inputs/map.xodr" -p project_absolute_path:="$PWD" -p gpu_support:=true
To run without GPU support, disable the gpu_support toggle
ros2 run sumoware scenario_controller --ros-args -p use_sim_time:=true -p scenario_file:="./example_inputs/scenario_params.json" -p map_file:="./example_inputs/map.xodr" -p project_absolute_path:="$PWD" -p gpu_support:=false
For a step-by-step installation guide please click here
Simulation Videos
Videos of simulations running SumoWare can be found here.
Contributors
Faruk Öztürk
Evald Nexhipi
Mathias Pechinger
Paper
If you use this repository, please cite our paper:
Öztürk, F., Nexhipi, E., Pechinger, M., & Bogenberger, K. (2025). SumoWare: Bridging SUMO and Autoware to Assess AV-Induced Traffic Impact. IEEE IV 2025, 2415–2421. DOI.
BibTeX:
@inproceedings{sumoware2025, title={SumoWare: Bridging SUMO and Autoware to Assess AV-Induced Traffic Impact}, author={{\"O}zt{\"u}rk, Faruk and Nexhipi, Evald and Pechinger, Mathias and Bogenberger, Klaus}, booktitle={2025 IEEE Intelligent Vehicles Symposium (IV)}, pages={2415--2421}, year={2025}, organization={IEEE}, doi={10.1109/IV64158.2025.11097507} }
Disclaimer
This project contains code derived from Autoware and tier4_autoware_msgs, licensed under the Apache License 2.0.