DANCERS-docker

Docker containers to install and run the DANCERS co-simulator.

Install

We provide the Dockerfile(s) to install the DANCERS co-simulator inside a container.

There is 3 environments in this repository : env_viragh_ns3, env_gazebo_ns3, env_minidancers_ns3. They correspond to three different physics simulators that can be used with DANCERS. If it is your first time using DANCERS, we recommend using env_minidancers_ns3.

1. The first step is to build a docker image from the provided Dockerfile. Example : docker build <environment-name> -t <image-name> --build-arg USERNAME=teatime This step requires an internet connexion and can take several minutes.
2. Authorize docker to open windows on your host's X server: xhost +local:docker
3. Once generated, you can create a docker container and connect to it with the following command (explained below): docker run --rm -it -e "TERM=xterm-256color" -e "DISPLAY=$DISPLAY" -v "/tmp/.X11-unix:/tmp/.X11-unix" -v "$HOME/.Xauthority:/root/.Xauthority:ro" --network host <image-name> --rm destroys the container at the end of utilization, kind of hazardous if you quit the container unexpetedly, you can remove it safely
   -it is for interactive, it opens a shell inside the container once it runs
   -e "TERM=xterm-256color" defines the terminal to be used in docker (colours, yey !)
   -e "DISPLAY=$DISPLAY" is necessary to let the container know to which display it should open windows
   -v "/tmp/.X11-unix:/tmp/.X11-unix" mounts a volume to the X server running in the host (allows the container to find your X server)
   -v "$HOME/.Xauthority:/root/.Xauthority:ro" Give the container access to create windows with the X server of the host
   --network host gives the container access to the host's networking
   <image-name> is the name of the image that you want to run (e.g. viragh_ns3_img)

Using the Mini-dancers variation

To launch your first simulation with our lightweight physics simulator, mini-dancers, build and run the mini-dancers environment: docker build env_minidancers_ns3 -t minidancers_ns3_img --build-arg USERNAME=teatime

Then, run this command from the DANCERS-docker repository: docker run --rm -it -e "TERM=xterm-256color" -e "DISPLAY=$DISPLAY" -v "/tmp/.X11-unix:/tmp/.X11-unix" -v "$HOME/.Xauthority:/root/.Xauthority:ro" -v "./env_minidancers_ns3/visualizer_v0.0.rviz:/home/teatime/.rviz2/visualizer_v0.0.rviz" --network host minidancers_ns3_img Once you are inside the container, run the tmux script that starts the simulation: cd sim_ws ./src/tmux_scripts/start_dancers_minidancers.sh src/config/config_minidancers_default.yaml ~/sim_ws/ A Rviz2 window should open and show 6 agents performing "string-like" flocking between two targets in a corridor environment:

Using the Viragh variation

To launch your first simulation, run the following tmux script from the container: ./sim_ws/src/tmux_scripts/start_dancers_viragh.sh ~/flocking I opens a tmux session with several terminals and the windows of the Viragh's simulator GUI. The tmux session: top-left : Viragh's simulator
top-right : DANCERS's Coordinator module
bottom-left : ns-3
bottom-right : Connector module for Viragh's simulator

Note: The tmux prefix has been changed to Ctrl - a (instead of default Ctrl - b)

Press spacebar to launch the simulation.

To quit the simulation, press Ctrl-a and ':' to access tmux CLI and type 'kill-session' then press enter.