Halation

The Stanford Antarctic Rover project.

Dependencies:

• Orbstack/Docker

Setup

1. docker-compose up -d
2. docker-compose exec halation bash
3. colcon build --symlink-install ALWAYS RUN FROM THE ROOT (not src!)!
4. source install/setup.bash
5. On subsequent starts, run 1-2 (no need to build or source install)

To rebuild the docker image, run docker-compose up --build -d

Development

• You can directly just edit the code if you have ROS2 already setup on your system, but most likely you won't be.
• If you don't have the ROS2 setup, then you won't have autocomplete/typedefs/important things that you probably want to develop with.
• Instead, we've set up two ways you can access the code.
  • (untested) There is a devcontainers.json, which should allow you to directly develop within the container using an VSCode extension 'Dev Containers' (somehow doesn't work with Hinson's machine)
  • I've additionally set up ssh, so that if devcontainers doesn't work for you, you can use 'Remote-SSH' (tested on Cursor) with ssh root@localhost -p 2267. The password is set in the Dockerfile as thispasswordissecure, although since this is just local ssh, you can do whatever you want.
• Either of these methods should get you the proper type definitions. If not, make sure the interpreter is set to /usr/bin/python3; if it still doesn't work, ping Hinson ;)

Execution

• ros2 launch rover_bringup rover_system.launch.py

SSHing into the system

• Install cloudflared (brew install cloudflared for Macs, follow .deb package + dpkg -i ...deb for Linux)
• Follow ssh-keygen instructions to generate ssh key (give to Hinson)
• Paste this in your ~/.ssh/config

  Host jetson.ssirovers.org
      ProxyCommand cloudflared access ssh --hostname %h
      User jetson
  

• Run ssh jetson.ssirovers.org

LIDAR / Object Avoidance / Rerouting Execution

• ROS package for object avoidance/rerouting using LIDAR is contained within src/avoidance_rerouting
• Launch using: ros2 run avoidance_rerouting mock_lidar_publisher (this starts a node that sends mock LIDAR signals to /scan)
• To visualize LIDAR outputs, run: ros2 launch foxglove_bridge foxglove_bridge_launch.xml, then go to https://app.foxglove.dev/ssi-halation/dashboard (ask Hiroki to be invited to dashboard) and open up ws://localhost:8765 on Foxglove.

External ROS2 Packages related to LIDAR / Object Avoidance / Rerouting Execution

• laser_segmentation: Convert raw LIDAR data to object segementation "https://github.com/ajtudela/laser_segmentation"
• slg_msgs: Dependency of laser_segmentation, used to process laser data "https://github.com/ajtudela/laser_segmentation"

Dependencies

• We use uv to manage the python dependencies, but NOT for project running. That is, uv add and ./uvsync.sh.
• Disregard the .venv/, ros itself has it's own types and deps we need.

TODO:

• Figure out proper configuration for environment variables
• How to do shutdowns

On old Ubuntu versions/Jetson

docker compose up -d docker compose exec halation bash