CS Freiburg: Doing the Right Thing in a Group

Abstract

Abstract. The success of CS Freiburg at RoboCup 2000 can be attributed to an effective cooperation between players based on sophisticated soccer skills and a robust and accurate self-localization method. In this paper, we present our multiagent coordination approach for both, action and perception, and our rich set of basic skills which allow to respond to a large range of situations in an appropriate way. Furthermore our action selection method based on an extension to behavior networks is described. Results including statistics from CS Freiburg final games at RoboCup 2000 are presented.

A Formal Analysis and Taxonomy of Task Allocation in Multi-Robot Systems

“…The resulting coordination problem can be cast as an iterated assignment problem in which the robots'roles are periodically reevaluated, usually at a frequency of the order of 10 Hz. This utility-based dynamic role assignment problem has been studied by many (Stone and Veloso 1999;Castelpietra et al 2001;Weigel et al 2001;Emery, Sikorski, and Balch 2002;Vail and Veloso 2003).…”

The Evaluation Of Performance

“…Ball possession is particularly relevant for the physical agent challenge in contrast to the simulator league because physical components (movement) and vision and sensor fusion (localization) are an integral part of this challenge [2]. The different categories of ball possession are constituted by relating them to the existing repertoire of actions of the CS Freiburg team [4] and assuming that the best possible action which may conform to the observed behavior is executed. In the following, we will distinguish between different kinds of (non-) ball possession: -ball free: no player is at the ball.…”

Cooperating Physical Robots: A Lesson in Playing Robotic Soccer

“…The players of the CS Freiburg team [11,18,23] use laser range finders in order to solve the self-localization problem [12], and for this reason can be assumed to know their own position very reliably. However, they are not very good in recognizing the ball and estimating its position on the field -which is done using a monocular vision camera.…”