A popular approach to describing the spatial environment of an autonomous system is to compute a representation for the local space. We have developed a theory of cognitive mapping that is based on the idea of an autonomous agent computing a network of such representations for the place it visits. In this paper we show a robot is able to compute the local space representation from the initial view it has of its environment The robot literally "takes it first step" to compute its first local environment using our theory of cognitive mapping.
展开▼
