What is it about?

A soccer robot plays on a field, which always looks the same. Even the own and opponent goal look the same, while the distinction is essential to win a game. By learning color patterns in the surroundings, a soccer robot can estimate reliably its heading.

Featured Image

Why is it important?

A magnetic compass is not always available and the estimates of an inertia sensor slowly drift. The optical compass described in this article works for every robot with a camera, as long as enough detail is visible in the surroundings.


This is impressive work, because first is learned what the distinctive colors in the background are, before a polar map of the surroundings is made. In addition, this approach is demonstrated not only on the small field inside our lab, but also outside on a full scale human soccer field.

Dr. Arnoud Visser
Universiteit van Amsterdam

Read the Original

This page is a summary of: An appearance-based visual compass for mobile robots, Robotics and Autonomous Systems, May 2009, Elsevier,
DOI: 10.1016/j.robot.2008.10.002.
You can read the full text:



The following have contributed to this page