Task decomposition and dynamic role assignment for real-time strategic teamwork

Abstract

Multi-agent domains consisting of teams of agents that need to collaborate in an adversarial environment offer challenging research opportunities. In this paper, we introduce periodic team synchronization domains, as time-critical environments in which agents act autonomously with limited communication, but they can periodically synchronize in a full-communication setting. We present a team agent structure that allows for an agent to capture and reason about team agreements. We achieve collaboration between agents through the introduction of formations. A formation decomposes the task space defining a set of roles. Homogeneous agents can flexibly switch roles within formations, and agents can change formations dynamically, according to pre-defined triggers to be evaluated at run-time. This flexibility increases the performance of the overall team. Our team structure further includes pre-planning for frequent situations. We fully implemented this approach in the domain of robotic soccer. Our simulator team made it to the semi-finals of the RoboCup-97 competition, in which 29 teams participated. It achieved a total score of 67-9 over six different games, and successfully demonstrated its flexible team structure. Using the same team structure, our small robot team won the RoboCup-97 small-robot competition, in which 4 teams participated. It achieved a total score of 13-1 over 4 games and also demonstrated its flexible team structure.