A cognitive map for an artificial agent

Abstract

We show how a general-purpose cognitive architecture augmented with a general diagrammatic component can represent and reason about Large-scale Space. The diagrammatic component allows an agent built in this architecture to represent information both symbolically and diagrammatically as appropriate. Using examples we show (a) how the agent's bimodal representation captures its knowledge about large-scale space as well as how it learns this information while problem solving and (b) the agent's flexibility when it comes to using learned information and incorporating new information in solving problems involving large-scale space.