| Week | Main Topic | Sub Topic(s) |
|---|---|---|
| Day 1 | Intro | Introduction to Robotics, Introduction to ROS2 |
| Day 2 | Computer Vision | Computer Vision, Project |
| Day 3 | LiDAR | LiDAR, Project |
| Day 4-5 | Workshop & Project |
Planning: Link
Repository: https://github.com/PXLAIRobotics/ROS2_Development_Environment
Invite link: https://discord.gg/ftrxEKkS
- 01 - Introduction to ROS
- 02 - ROS Concepts
- 03 - ROS: Publisher Subscriber
- 04 - Simulation Environments
- Start the gazebo sim
gz sim
...and start the Prius on Sonoma Raceway world.
- Create a ROS bridge for the
/cmd_veltopic
ros2 run ros_gz_bridge parameter_bridge /cmd_vel@geometry_msgs/msg/Twist@gz.msgs.Twist
-
Create a Teleop Node that allows you to control the Prius using keyboard inputs.
-
Always remember to rebuild and source each time you change anything:
clean_and_rebuild_ros2_workspace
If that does not work, you can do it manually:
colcon build
source install/setup.bash
- Go nuts
- Start the gazebo sim
gz sim
...and start the Prius on Sonoma Raceway world.
- Create a ROS bridge for the
/front_cameratopic
ros2 run ros_gz_bridge parameter_bridge /front_camera@sensor_msgs/msg/Image@gz.msgs.Image
- Create a vision_controller package:
cd ~/Projects/ros2_workspace/src
ros2 pkg create --build-type ament_python vision_controller --dependencies rclpy image_transport cv_bridge sensor_msgs std_msgs opencv2
-
Add the
src/process_front_camera.pyscript -
Always remember to rebuild and source each time you change anything:
clean_and_rebuild_ros2_workspace
If that does not work, you can do it manually:
colcon build
source install/setup.bash
- Drive autonomously along the racetrack. Use the Example OpenCV code for inspiration.
It's recommended to use a Miro board for your business idea (https://miro.com/) to collaborate easily.
Technology path
You can show code examples, but limit it to the most relevant sections, i.e.:
- Usage of a canny edge detector
- Contour detection
Focus on visuals, i.e:
- Screenshots or video of the results of computer vision transformations
- Live demo of the car driving
- ...
Innovation path
- Use the AI Business model canvas as a starting point, but you can convert it into a presentation. Your choice. Make it a striking presentation!
How to get an overview of all gazebo topics:
gz topic -l
Then you can use a ROS-Gazebo bridge for whatever topic you need.
- ROS2 (Jazzy): https://docs.ros.org/en/jazzy/Tutorials.html
- Gazebo: https://gazebosim.org/docs/harmonic/tutorials/
- Gazebo x ROS2: https://github.com/gazebosim/ros_gz/tree/ros2/ros_gz_bridge