What is Docker?

Docker is an open-source platform that lets you package software and all of its dependencies into a self-contained unit called a container. A container bundles the application code, runtime, system libraries, and configuration files needed to run a piece of software—regardless of the host machine’s operating system or installed packages.

Containers vs. Virtual Machines

Both containers and virtual machines (VMs) provide isolation, but they differ fundamentally in how they work:

Feature	Container	Virtual Machine
Boot time	Milliseconds	Minutes
Disk footprint	Tens of MB	Several GB
OS kernel	Shared with host	Full guest OS
Isolation level	Process-level	Hardware-level
Performance	Near-native	Overhead from hypervisor

Containers share the host operating system kernel, which makes them much lighter and faster than VMs. Each container still has its own filesystem, network stack, and process space, so it remains isolated from everything else running on the machine.

Key Concepts

Before working with Docker, it is important to understand a few fundamental terms:

Image : A read-only template that defines the container’s contents—operating system, installed packages, environment variables, and startup commands. Think of it as a snapshot or a recipe. Images are immutable; you never modify them directly.

Container : A running instance of an image. You can start many containers from the same image, and each one is isolated from the others. Containers are ephemeral by default—any changes made inside are lost when the container is stopped or removed, unless you explicitly persist them.

Registry : A server that stores and distributes Docker images. The public registry is Docker Hub. PAL Robotics uses the GitLab Container Registry to distribute its images privately.

Volume / Bind mount : A mechanism to share a folder between the host machine and a running container. This is how you persist files across container restarts and share data with your host environment.

Dockerfile : A text file containing the instructions needed to build a Docker image from scratch (or from an existing base image).

Why Docker for Robotics?

Working with a robot like Kangaroo involves a specific combination of ROS 2 packages, simulation tools, controllers, and system libraries that must be compatible with each other and with the robot firmware. Setting up this environment manually on every developer machine is error-prone and time-consuming.

Docker solves this by:

• Guaranteeing reproducibility: every developer and CI pipeline uses exactly the same environment.

• Eliminating “works on my machine” issues: the container is the environment.

• Enabling fast onboarding: a new team member can be up and running in minutes.

• Keeping the host clean: all ROS packages, build tools, and simulation binaries live inside the container and do not interfere with other projects on your laptop.

Containerization Key Properties

Property	What it means in practice
Flexible	Even complex multi-package ROS workspaces can be containerized
Lightweight	Shares the host kernel — no full OS overhead
Interchangeable	Swap in a newer image version without touching your host
Portable	Build once, run on any Linux machine with Docker installed
Scalable	Spin up multiple containers for parallel CI jobs or multi-robot tests
Stackable	Compose multiple services (robot driver, simulation, visualizer) together

Further Reading

If you want to deepen your understanding of Docker, the official documentation is the best starting point:

• Docker overview

• Docker concepts — images & containers

• Docker volumes

• Dockerfile reference