An infrastructure for adaptive control of multi-agent systems

Abstract
We present the control infrastructure of the Cougaar distributed agent system that was developed under the DARPA ALP and UltraLog programs. We motivate its design from a control theory perspective and discuss the characteristics of an agent system as the controlled process. These characteristics are arguably the reason why formal methods of control theory are rarely applied in software engineering for agent systems.

This publication has 6 references indexed in Scilit: