On Being a Teammate: Experiences Acquired in the design of RoboCup Teams

Increasingly, multi-agent systems are being designed for a variety of complex, dynamic domains. E ective agent interactions in such domains raise some of most fundamental research challenges for agentbased systems, in teamwork, multi-agent learning and agent modeling. The RoboCup research initiative, particularly the simulation league of RoboCup, has been proposed to pursue such multi-agent research challenges, using the common testbed of simulation soccer. However, despite the signi cant popularity of RoboCup, researchers have often not extracted the general lessons learned from their participations in RoboCup. This is what we attempt to do here. We have elded two teams, ISIS97 and ISIS98, in RoboCup competitions. These teams have been in the top four teamsin these competitions. We compare the teams, and attempt to analyze and generalize the lessons learned. This analysis reveals several surprises, pointing out lessons for teamwork and for multi-agent learning. 1 

• Type: Conference
• Year: 1999
• Where: AGENTS
• Authors: Stacy Marsella, Jafar Adibi, Yaser Al-Onaizan, Gal A. Kaminka, Ion Muslea, Milind Tambe
• Keywords: AGENTS 1999, Multi-agent, Multi-agent Learning, Research Challenges, Security Privacy