| Ubuntu | ROS 2 | Autoware | Nvidia Driver | CUDA Version |
|---|---|---|---|---|
| 20.04 | galactic | auto/(source installation) | 470.63.01 | 11.4+ |
| CPU | RAM | GPU |
|---|---|---|
| 8 cores | 16 GB | NVIDIA 4GB RAM |
GPU is not required to run basic functionality, it is mandatory to enable the following neural network related functions:
- LiDAR based object detection
- Camera based object detection
- Traffic light detection and classification
- Autoware.Auto source installation (https://autowarefoundation.gitlab.io/autoware.auto/AutowareAuto/installation-no-ade.html)
- CARLA 0.9.13 source installation (debian, https://carla.readthedocs.io/en/latest/start_quickstart/)
- carla-auto-bridge (this repository)
mkdir -p ros2_ws/src
cd ros2_ws/src
git clone [email protected]:CL2-UWaterloo/carla-auto-bridge.git --recursive- Download CARLA maps (https://bitbucket.org/carla-simulator/autoware-contents/src/master/maps/), name accordingly
point_cloud/Town01.pcd -> Town01/pointcloud_map.pcd
vector_maps/lanelet2/Town01.osm -> Town01/lanelet2_map.osm
cd ros2_ws
colcon build --symlink-install- Run CARLA, load a map
/opt/carla-simulator/CarlaUE4.sh
python3 /opt/carla-simulator/PythonAPI/util/config.py -m Town01- Launch carla_ros_bridge, autoware.auto bridge
ros2 launch carla_ros_bridge carla_ros_bridge_with_example_ego_vehicle.launch.py
ros2 launch autoware_bridge autoware_bridge_launch.xml
ros2 launch autoware_bridge autoware_launch.xml map_path:=<path-to-map-dir>/Town01 vehicle_model:=sample_vehicle sensor_model:=sample_sensor_kit- Set initial pose
- Set goal position
- Wait for planning
- Engage (use web interface or Rviz plugin (AutowareStatePanel))