Using norms to control open multi-agent systems

Abstract

In this thesis we propose to use norms, which are formal descriptions of acceptable behaviours, to define control, coordination and cooperation mechanisms for multi-agent systems. Specifically, in this thesis we have developed norm-based mechanisms for open multi-agent systems at two levels: at agent level we address the problem of defining norm-autonomous agents that deliberate about norms within uncertain environments; and at multi-agent system level we propose a distributed architecture for enforcing norms in open multi-agent systems.