An Emergent Framework For Self-Motivation In Developmental Robotics

Abstract

This paper explores a philosophy and connectionist al-gorithm for creating a long-term, self-organizing develop-mental robot control system. This intrinsic algorithm and architecture implements self-motivation by creating a sys-tem capable of anticipating its next state, while simultane-ously attempting to seek out that for which it cannot predict. These competing internal pressures are designed to drive the system in a manner reminiscent of a co-evolutionary arms race. 1.