Towards a framework for robot cognition

Abstract

This work describes a framework for control of attention and for pattern categorization using a robot platform consisting of an articulated stereo-head with four degrees of freedom (pan, tilt, left verge, and right verge). As a practical result of this work, the system can select a region of interest, perform attention shifts involving saccadic movements, perform efficient feature extraction and recognition, incrementally construct a world map, and keep the map consistent with a current perception of the world. Another important result for the attentional mechanism is that the system is capable of analyzing all regions of its world, selected according to a salience map.