Applying computational theories of cognitive mapping to mobile robots

Abstract

Artificial intelligence researchers have produced many computational theories of cognitive mapping. However, few of their ideas have been adapted to mobile robots. This paper outlines a new computational theory of cognitive mapping and then describes an attempt to implement the main properties of the theory on an actual mobile robot. Naturally, the theory needs to be modified to accommodate the strengths and weaknesses of a mobile robot. Preliminary experiments were conducted with the mobile robot and results are given.