Abstract

This work describes a control architecture based on a hierarchical classifier system. This system, which learns both reactive and planning rules, implements a motivationally autonomous animat that chooses the actions it performs according to its perception of the external environment, to its physiological or internal state, to the consequences of its current behavior, and to the expected consequences of its future behavior. The adaptive faculties of this architecture are illustrated within the context of a navigation task, through various experiments with a simulated and a real robot. I. Introduction The work presented in this paper fits into the so-called animat approach, which aims at designing animats, i.e., simulated animals or real robots whose rules of behavior are inspired by those of animals. The proximate goal of this approach is to discover architectures or working principles that allow an animal or a robot to exhibit an adaptive behavior and, thus, to survive or fulfill i...

Citations