In vision-based robotic systems the robust tracking of scene features is a key element of grasping, navigation and interpretation tasks. The stability of feature initialisation and tracking is strongly influenced by ambient conditions, like lighting and background, and their changes over time. This work presents how robustness can be increased especially in complex scenes by reacting to a measurement of the scene content. Element candidates are proposed, to indicate the scene complexity remaining after running a method. Local cue integration and global topological constraints are applied to select the best feature set. Experiments show in particular the success of the approach to disambiguate features in complex scenes.
展开▼
