Peter Stone & Manuela Veloso
Robotic soccer is a multi-agent domain consisting of teams of agents that need to collaborate in a real-time, noisy, adversarial environment. As such, this exciting domain provides a myriad of AI research opportunities related to multi-agent systems, machine learning, real-time planning, opponent modeling, intelligent robotics, and several other AI topics.
At IJCAI-97, conference attendees had the opportunity to witness the first robotic soccer world cup (RoboCup) competitions. Another competition will be held at IJCAI-99. Competitions are held within both simulation and real platforms. This tutorial is an opportunity to learn the full range of research that has been going on behind the competitions.
This tutorial will be of interest to AI researchers and practitioners concerned with real-time multi-agent systems, including entertainment domains. The tutorial will also be particularly relevant to current and potential robotics researchers. It will help summarize the current state of the art in a coherent framework so that they can build upon past accomplishments and identify future research opportunities.
The tutorial will specifically introduce the RoboCup simulator and robotic platforms and the algorithms developed for multi-agent control learning, teamwork architecture, and distributed robotic perception and action. In addition, we plan to take special care to identify the generality of the techniques developed within RoboCup for researchers interested in other domains similar to robotic soccer.
Prerequisite knowledge:
Participants in this tutorial need not have previous familiarity with
robotic soccer.
Peter Stone
is a Ph.D: candidate in Computer Science Department at Carnegie Mellon
University. He received his M.Sc. in Computer Science from Carnegie
Mellon University in 1995 and his B.Sc. in Mathematics from the
University of Chicago in 1993. His research interests include planning
and machine learning, particularly in multi-agent systems. Mr Stone
has been a central figure in the creation of the RoboCup initiative,
currently serving as the chair of the RoboCup-99 simulator committee.
E-mail: pstone@cs.cmu.edu
Homepage: http://www.cs.cmu.edu/~pstone/
Manuela M. Veloso is Associate Professor of Computer Science at Carnegie Mellon University. She received her Ph.D. in Computer Science from Carnegie Mellon University in 1992. Prof. Veloso researches in the area of artificial intelligence.
Her long-term research goal is the effective construction of systems
of multiple agents where cognition, perception, and action are
combined to address planning, execution, and learning tasks. Prof
Veloso is the U.S. representative and founding member of the
International Committee for the RoboCup International Federation.
E-mail: veloso@cs.cmu.edu