The SocRob Project: Soccer Robots or Society of Robots

Abstract

Cooperative Robotics is a modern research field, with applications to areas such as building surveillance, transportation of large objects, air and underwater pollution monitoring, forest fire detection, transportation systems, or search and rescue after largescale disasters (Balch, T. & Parker, L., 2002). In short, a population of cooperative robots behaves like a distributed robot to accomplish tasks that would be difficult, if not impossible, for a single robot. Many lessons important for this domain can be learned from the Multi-Agent Systems field of Artificial Intelligence (AI) concerning relevant topics for Cooperative Robotics, such as distributed continual planning (desJardins, M. E., et al, 1999), task allocation (Ferber, J., 1999), communication languages or coordination mechanisms (Decker, K. S., & Lesser, V. R., 1995). Robotic soccer is a very challenging problem, where the robots must cooperate not only to push and/or kick an object (a ball) towards a target region (the goal), but also to detect and avoid static (walls, stopped robots) and dynamic (moving robots) obstacles while moving towards, moving with or following the ball. Furthermore, they must cooperate to defeat an opposing team. All these are features common to many other cooperative robotics problems. This paper surveys the several research problems addressed by the SocRob project, carried out by the Intelligent Systems Laboratory at the Institute for Systems and Robotics Instituto Superior Tecnico (ISR/IST) in Lisbon, building a Systems Theory standpoint on AI concepts. In Section 2, we describe our view of the general problem involving multiple robots that act as a team, cooperating and coordinating their actions to attain the team goal. Needless to say, single-robot ''traditional'' research problems are covered, both from the sub-system and from the integration standpoints. Natural extensions to cooperative multi-robot teams are also detailed. The problems addressed so far and the solutions we obtained for them are described in Section 3. Open problems of interest for the project and clues on how we intend to approach their solution are discussed in Section 4. We end the paper drawing some conclusions in Section 5.