Installing ROS 2 on macOS with Docker

Many native tools in the ROS ecosystem are only supported on Linux, often making robotics development on other operating systems difficult and tedious. To solve this developer friction, many roboticists have turned to container platforms like Docker to help them work cross-platform.

In this tutorial, we'll cover how you can quickly get up and running with ROS 2 on your macOS computer – all with the help of Docker containers.

Pulling a Docker image

Docker is a container platform that allows you to run an instance of an operating system on top of a host. It can empower teams to develop robots with unique operating system requirements, regardless of their own system specs, or to work with ROS distributions that are not supported on the host operating system.

A Docker image essentially packages software into standardized units for reproducible development and deployment. It's a lightweight, standalone, executable piece of software that includes everything needed to run an application – all the code, runtime, system tools, system libraries and settings.

Once you've installed Docker on your macOS machine, you can browse Docker Hub for available ROS images. Download (pull) your desired image (in this case, ROS 2 Jazzy) from Docker Hub:

$ docker pull ros:jazzy-ros-core


language-bash

Verify that you've downloaded the correct image:

$ docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
ros jazzy-ros-core cbe7e6024f69 8 weeks ago 505MB


language-bash

Customize the container

Before we run any commands, we need to customize the image by installing a few more packages. To do so, create a file named Dockerfile in an empty directory with the following contents:

FROM ros:jazzy-ros-core
RUN apt-get update && apt-get install -y \
ros-jazzy-demo-nodes-cpp \
ros-jazzy-foxglove-bridge \
ros-jazzy-tf2-ros


language-dockerfile

In a new terminal, run the following command to build the image and tag it with the name rosdemo:

$ docker build -t rosdemo .


language-bash

The -t option tags the new image with the name rosdemo. The . represents the build context, which is the current folder.

Running the container

You can now start your ROS 2 container:

$ docker run --rm -it rosdemo bash


language-bash

The --rm flag will remove this container after it exits (to save disk space), and the -it option will open an interactive terminal.

Publishing data in the container

Inside your docker container, publish a /chatter topic:

root@396119f9e9ce:/# ros2 run demo_nodes_cpp talker


language-bash

In another terminal, start a new docker container and verify you can see the message being published. This time, instead of running Docker with bash we're going to run the ros2 command directly:

$ docker run --rm -it rosdemo ros2 topic echo /chatter


language-bash

Your output should look similar to the following. Keep in mind that both containers are fully isolated, and are talking to each other over Docker's internal network.

Visualizing the data in Foxglove

Next, let's verify our ROS messages by visualizing them in Foxglove.

To be able to connect Foxglove to the container, you must publish the port 8765 on the Docker server with the -p 8765:8765 option when running the new container. Start a third docker container, this time launching the Foxglove ROS Bridge:

docker run --rm -it -p 8765:8765 rosdemo ros2 launch foxglove_bridge foxglove_bridge_launch.xml


language-bash

To make things more interesting, let's also publish a simple transform:

$ docker run --rm -it rosdemo ros2 run tf2_ros static_transform_publisher --x 0 --y 1 --z 1 --roll 0 --pitch 0 --yaw 0 --frame-id base_link --child-frame-id sensor


language-bash

Now, open your Foxglove desktop app, initiate a Foxglove WebSocket connection with the default URL. You can also use the following direct link:

https://app.foxglove.dev/~/view?ds=foxglove-websocket&ds.url=ws%3A%2F%2Flocalhost%3A8765

Add a Raw Messages panel and subscribe to the /chatter topic to verify that it shows the message "hello":

You've now successfully run ROS 2 in a Docker container, and connected it to Foxglove to visualize that data.

Connect with us

With Docker containers, you can seamlessly continue development across different platforms – regardless of your robotics framework's constraints.

Reach out to us directly in our Discord community to ask questions or request a topic for the next tutorial.