Building agent teams using an explicit teamwork model and learning

Multi-agent collaboration or teamwork and learning are two critical researc h challenges in a large number of multi-agent applications. These research challenges are highlighted in RoboCup, an international project focused on robotic and synthetic soccer as a common testbed for research in multi-agen t systems. This article describes our approach to address these challenges, based on a team of soccer-playing agents built for the simulation league o f RoboCup-the most popular of the RoboCup leagues so far. To address the challenge of teamwork, we investigate a novel approach based on the (re)use of a domain-independent, explicit model of teamwork, an exp licitly represented hierarchy of team plans and goals, and a team organizat ion hierarchy based on roles and role-relationships. This general approach to teamwork, shown to be applicable in other domains beyond RoboCup, both r educes development time and improves teamwork flexibility. We also demonstr ate the application of off-line and on-line learning to improve and special ize agents' individual skills in RoboCup. These capabilities enabled our so ccer-playing team, ISIS, to successfully participate in the first internati onal RoboCup soccer tournament (RoboCup'97) held in Nagoya, Japan, in Augus t 1997. ISIS won the third-place prize in over 30 teams that participated i n the simulation league. (C) 1999 Elsevier Science B.V. All rights reserved .