What is it about?
A camera recording warps a 3D world onto a 2D image, a transformation which can be estimated when enough information is available. Yet, there are situations when there are only two landmarks visible in the point of view, which means that there are multiple positions which could have given the same image. Because a robot like the Nao has an inertia sensor in his body, the gravitational direction is known, which allows to estimate the reduce the problem to a single solution.
Featured Image
Photo by Marcel Eberle on Unsplash
Why is it important?
When there is enough texture around your robot, there are enough points to estimate your position, as Structure for Motion has proven. Yet, there are many situations (for instance after a disaster), when you there are only a few recognizable landmarks available, and we also need a solution for that situation.
Perspectives
This is a fundamental problem, where many scientist have worked on, and still do. For instance, I am very proud that we could fall back to a publication of Finstenvalder and Scheufele from 1903, which cited on their turn Snellius (1617). That really gives the feeling that you are standing on the shoulders of giants.
Dr. Arnoud Visser
Universiteit van Amsterdam
Read the Original
This page is a summary of: Position and Altitude of the Nao Camera Head from Two Points on the Soccer Field Plus the Gravitational Direction, January 2025, Springer Science + Business Media,
DOI: 10.1007/978-3-031-85859-8_14.
You can read the full text:
Contributors
The following have contributed to this page
