VIP neuron model: Head-centered cross-modal representation of the peri-personal space around the face

摘要

Since body representation is one of the most fundamental issues for physical agents (humans, primates, and also robots) to adaptively perform various kinds of tasks, a number of learning methods have attempted to make robots acquire their body representation. However, these previous methods have supposed that the reference frame is given and fixed a priori. Therefore, such acquisition has not been dealt. This paper presents a model that enables a robot to acquire cross-modal representation of its face based on VIP neurons whose function (found in neuroscience) is not only to code the location of visual stimuli in the head-centered reference frame and but also to connect visual and tactile sensations. Preliminary simulation results are shown and future issues are discussed.