Skip to content

Week 3 · ROS 2 (Humble) with Python

This week you’ll learn how ROS 2 is organized and practice the core ideas you’ll use all semester: workspaces, packages, publish/subscribe, and a tiny sim (turtlesim). The steps below point you to the official Humble tutorials—follow them carefully.

Part A - Setup

First, we'll show you how to make shortcut commands to launch your Docker image. The commands you need to run will vary depending on wether or not you are using the Nvidia container.

For people not using the Nvidia container, run: 

echo -e "#"'!'"/bin/bash\nexport userid=$(id -u) groupid=$(id -g)\ncd ~/ENME480_mrc/docker\ndocker compose -f humble-enme480_ur3e-compose.yml run --rm enme480_ur3e-docker" > startDocker.sh

echo -e "#"'!'"/bin/bash\ncontainer="'$(docker ps | grep docker-enme480_ur3e-docker-run | cut -b 1-12)'"\necho Found running container "'$container'". Connecting...\ndocker exec -ti "'$container'" bash" > connectToDocker.sh

For people who are using the Nvidia container, run: 

echo -e "#"'!'"/bin/bash\nexport userid=$(id -u) groupid=$(id -g)\ncd ~/ENME480_mrc/docker\ndocker compose -f humble-enme480_ur3e-nvidia-compose.yml run --rm enme480_ur3e-docker" > startDocker.sh

echo -e "#"'!'"/bin/bash\ncontainer="'$(docker ps | grep docker-enme480_ur3e-docker-run | cut -b 1-12)'"\necho Found running container "'$container'". Connecting...\ndocker exec -ti "'$container'" bash" > connectToDocker.sh

These commands will create two bash files (essentially just lists of other commands) that will allow you to launch the Docker container with: 

bash startDocker.sh

And connect to it from another terminal with: 

bash connectToDocker.sh
Provided you are in the folder where these files are.

IMPORTANT: This the following folders will be linked (outside the docker --> inside the docker)

ENME480_mrc/src --> enme480_ws/src
ENME480_mrc/config --> enme480_ws/config
This is where you should place any files you want to keep when the docker shuts down (i.e. assignment code). Any changes made to files in these folders in the docker will be reflected outside the docker and vice versa. (Credit to Benjamin Ruby for the original version of the script)

🛫 Part B — Pre-flight check (5–10 min)

1) Using the commands from above, open your Docker image contianing ROS 2) Open a new terminal and verify ROS 2 is available (e.g., ros2 --version or ros2 --help).

ros2 topic list
It will show you a list of topics, most likely /rosout and /parameter_events

3) Try opening up Gazebo 

ign gazebo
It should open up a window with examples to different test environments

4) Try opening up rqt 

rqt
If rqt does not open up anything or throws an error, install any required extenstions 
sudo apt install ros-humble-rqt*
or try reopening it with 
rqt --force-discover

Troubleshooting

During any of these steps if your display doesn't open up, follow the follwoing steps

Checkpoint B (no submission yet): You can run the above commands without any errors.

📦 Part C — Workspace & package (setup only)

Follow the official Humble tutorials step-by-step (do not copy solution code from elsewhere):

1) Create a workspace: Use the Create a workspace guide and build once so the structure is valid.
↪ Guide: docs.ros.org/en/humble/Tutorials/Beginner-Client-Libraries/Creating-A-Workspace/Creating-A-Workspace.html

2) Create a Python package: Inside your src/, make a new package for this week (any sensible name, e.g., week3).
↪ Guide: docs.ros.org/en/humble/Tutorials/Beginner-Client-Libraries/Creating-Your-First-ROS2-Package.html

Checkpoint C: Your workspace builds with colcon and your package appears in the build output.

📡 Part D — Talker / Listener

Use the publisher/subscriber (Python) tutorial as your primary reference:

Task C1 — Publisher (“talker”)
Create a node that publishes numbers at a steady rate on a topic you choose (e.g., /numbers).
(Use the tutorial to recall how to create a publisher node; adapt it to publish numbers rather than strings.)

Task C2 — Subscriber (“listener”)
Create a node that subscribes to your numbers topic and maintains a cumulative sum. After each new message arrives, it should publish the running sum on a new topic (e.g., /sum_topic).
(Re-use the subscriber pattern from the tutorial; add your own sum logic and a second publisher.)

Task C3 — Inspect with CLI
Use the Understanding topics tutorial to: list topics, echo your sum topic, and show topic info (type, publishers/subscribers).
↪ docs.ros.org/en/humble/Tutorials/Beginner-CLI-Tools/Understanding-ROS2-Topics/Understanding-ROS2-Topics.html

Hints (conceptual, not solutions): - Topic names must match exactly; message types must be consistent.
- If a terminal shows “no publisher/subscriber”, confirm both nodes are running and your workspace is sourced.
- Keep node/topic names short and meaningful.

Checkpoint D — Screenshots to capture:
- Talker output (brief).
- Listener output showing a running sum.
- ros2 topic list and a short ros2 topic echo of your sum topic.

🐢 Part E — Turtlesim (drive in a circle)

Read the turtlesim, ros2, and rqt tutorial first:
docs.ros.org/en/humble/Tutorials/Beginner-CLI-Tools/Introducing-Turtlesim/Introducing-Turtlesim.html

Task D1 — Launch turtlesim
Start the turtlesim simulator in one terminal (see tutorial).

Task D2 — Velocity publisher
In your package, create a node that publishes velocity commands to turtlesim so the turtle moves in a circle (non-zero linear \(x\) and angular \(z\)).
(You’ll find the correct topic name in the tutorial; use the CLI to explore message fields.)

Task D3 — Observe & Inspect
Use ros2 topic list / ros2 topic echo to confirm motion and pose updates; open rqt and view the Node Graph and Topic Monitor.
- rqt info: docs.ros.org/en/humble/Concepts/Intermediate/About-RQt.html

Checkpoint E — Screenshots to capture:
- Turtlesim window with a clear circular path.
- Your velocity publisher terminal (brief output).
- ros2 topic list and a short ros2 topic echo of the pose topic.
- rqt with Node Graph / Topic Monitor visible.

🔎 What we’re assessing

  • You can follow official Humble docs and adapt examples to a new task.
  • Your nodes publish/subscribe correctly and use appropriate topic names and message types.
  • You can inspect systems via CLI and rqt and explain what you see.

📤 Deliverables (single PDF upload)

Include concise screenshots (one image may show multiple windows):

  • Talker/Listener: talker output; listener output showing running sum; ros2 topic list; brief ros2 topic echo of your sum topic.
  • Turtlesim: turtlesim showing a circle; velocity publisher terminal; ros2 topic list; brief ros2 topic echo of the pose topic; rqt with Node Graph/Topic Monitor visible.
  • A terminal view of your package tree (folder/files) helps grading.

Also submit the three Python files you created this week as separate attachments or in the PDF appendix (clearly named).

Helpful tips for the turtlesim velocity publisher

  • Topic to command motion: /turtle1/cmd_vel. That’s the velocity command topic turtlesim listens to. See the turtlesim tutorial for context.
  • Message type:
  • Make edits to package.xml to include geometry_msgs -geometry_msgs/Twist with two parts: linear and angular, each a 3-D vector (x, y, z). For a planar turtle, you’ll typically set linear.x and angular.z only (m/s and rad/s).
  • If you’re unsure of fields, ask ROS directly: ```bash ros2 interface show geometry_msgs/msg/Twist