IEEE International Conference on Robotics and Automation (ICRA), 2013
IEEE International Conference on Robotics and Automation location:Germany, Karlsruhe date:6-10 May 2013
This paper presents scene graphs for geometric relations between rigid bodies, using a semantic representation, invariant to the actual coordinate representation at hand, thereby making them more general then most scene graphs currently available. The presented scene graphs take into account constraints imposed by particular coordinate representations, allow for constant poses and handle semantic queries. Next to a pose graph, this paper presents a twist graph. The twist graph allows native twist calculations, as opposed to deriving the velocities from the pose graph. This paper comes with pointers to all software released under a dual BSD/LGPLv2.1 license.