Towards fast fiducial marker with full 6 DOF pose estimation
J. Ulrich, Ahmad Alsayed, Farshad Arvin, Tomáš Krajník
- Year
- 2022
- Citations
- 9
Abstract
This paper proposes a new method for the full 6 degrees of freedom pose estimation of a circular fiducial marker. This circular black-and-white planar marker provides a unique and versatile identification of individual markers while maintaining a real-time detection. Such a marker and the vision localisation system based on it is suitable for both external and self-localisation. Together with an off-the-shelf camera, the marker aims to provide a sufficient pose estimation accuracy to substitute the current high-end localisation systems. In order to assess the performance of our proposed marker system, we evaluate its capabilities against the current state-of-the-art methods in terms of their ability to estimate the 2D and 3D positions. For such purpose, a real-world dataset, inspired by typical applications in mobile and swarm robotics, was collected as the performance under the real conditions provides better insights into the method's potential than an artificially simulated environment. The experiments performed show that the method presented here achieved three times the accuracy of the marker it was derived from.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002