Inside Pavlov Mini Wheel (Part 2): From Camera Frame to Navigation Goal

ROS ROS Ideas
, , , ,
1

Hi everyone,

I’ve just published Part 2 of my Pavlov Mini Wheel series, where I take a deeper look at the software architecture behind my ROS 2 autonomous robot.

This article follows the complete perception pipeline—from a single camera frame to a navigation goal—and explains how the system is organized using independent ROS 2 nodes.

Topics covered include:

  • ROS 2 node architecture
  • Ball detection with OpenCV
  • 3D target localization
  • Publishing /ball/point and /ball/visible
  • Generating navigation goals with BallToGoal

The article also includes architecture diagrams, Gazebo screenshots, and code snippets from the actual project.

I’d love to hear your feedback or suggestions!

:open_book: Read the full article here:

https://medium.com/@bengokaysaglam/inside-pavlov-mini-wheel-building-an-autonomous-ros-2-robot-from-camera-to-motion-part-2-a09eca7967ae

The source code is available on GitHub: