Abstract
The identification of objects is a major function of vision, but provision of information about one's localization within a larger spatial environment is important as well (what is there vs. where am I). This task is of an inherently sensorimotor nature, i.e. results from a combination of sensory features and motor actions, where the latter comprise exploratory movements to different positions in the environment. To study the localization process we have designed an artificial agent which operates in virtual spatial environements. Its representation is realized in a hybrid architecture which integrates a bottom-up processing of sensorimotor features with a top-down reasoning that is based on the principle of maximum information gain. The system operates on two sensorimotor levels, a macro-level, which controls the movements of the agent in space, and a micro-level which controls its eye movements. As a result, the virtual mobile agent is able to orient itself within an environment using only a minimum number of exploratory actions. We evaluate its performance with a test suite consisting of different spatial environments and discuss the relation between the representation used by the agent and the properties of place cells in the hippocampus.
Supported by DFG SFB “Spatial Cognition”.