A SURVEY OF THE PROPERTIES OF AGENTS

Abstract

In the past decade agent systems were considered to be as one of the major fields of study in Artificial Intelligence (AI) field. Many different definitions of agents were presented and several different approaches describing agency can be distinguished. While some authors have tried to define “what” an agent really is, others have tried to identify agents by means of properties which they should possess. Most authors agree on these properties (at least basic set of properties) which are intrinsic to agents. Since agent's definitions are not consistent, we are going to give an overview and list the properties intrinsic to an agent. Many different adjectives were attached to the term agent as well and many different kinds of agents and different architectures emerged too. The aim of this paper it go give an overview of what was going on in the field while taking into consideration main streams and projects. We will also present some guidelines important when modelling agent systems and say something about security issues. Also, some existing problems which restrict the wider usage of agents will be mentioned too.