Home // ICAS 2011, The Seventh International Conference on Autonomic and Autonomous Systems // View article

A Simple Language for Describing Autonomous Agent Behaviors

Authors:
Philippe Codognet

Keywords: multi-agent systems, emergence, behavior languages, constraints, optimization

Abstract:
We present a simple and declarative language for describing autonomous behaviors in the paradigm of multi-agent systems. This framework is based on the notion of “goal constraints” and the satisfaction of these goals is done by iterative improvement, with the help of “repair functions” which will partially fulfill them, step by step. We can thus define both deterministic and non-deterministic behaviors. This framework is aimed at describing autonomous systems where the context is changing or unknown, thus goals can be re-evaluated or solved in different ways when changes occur. We will illustrate it with a simple example of life-like navigation behaviors for agents in unknown environments.

Pages: 105 to 110

Copyright: Copyright (c) IARIA, 2011

Publication date: May 22, 2011

Published in: conference

ISSN: 2308-3913

ISBN: 978-1-61208-134-2

Location: Venice/Mestre, Italy

Dates: from May 22, 2011 to May 27, 2011