Acquisition of qualitative spatial representation by visual observation

In robot navigation, one of the important and fundamental issues is to reconstruct positions of landmarks or vision sensors locating around the robot. This paper proposes a method for reconstructing qualitative positions of multiple vision sensors from qualitative information observed by the vision sensors, i.e., motion directions of moving objects. The process iterates the following steps: (1) observing motion directions of moving objects from the vision sensors, (2) classifying the vision sensors into spatially classified pairs, (3) acquiring three point constraints, and (4) propagating the constraints. The method have been evaluated with simulations.