Bayesioan estimation techniques, probabilistic relation models, higher-level knowledge representations and invariant motion descriptions for multi-target tracking and intent estimation in robotics. KU Leuven
The ability to keep track of targets, e.g. people or other robots, in real time, is fundamental in a wide range of robotic applications. Keeping track of targets however, is not sufficient in robotics. To allow for intelligent and safe robotic behavior the robot has to be aware, on top of the positions, of the intention of the targets. The goal of the proposed research is twofold: (1) multitarget tracking and localization (MTTL) and (2) target ...