AGILO RoboCuppers 2002: Applying Cooperative Game State Estimation Experience-based Learning, and Plan-based Control to Autonomous Robot Soccer

This paper describes the computational model underlying th e AGILO autonomous robot soccer team and its implementation. The mo st salient aspects of the AGILO control software are that it includes (1) a coope rative probabilistic game state estimator working with a simple off-the-shelf ca mera system; (2) a situated action selection module that makes amble use of exp eri nce-based learning and produces coherent team behavior even if inter-robot communication is perturbed; and (3) a playbook executor that can perform prep rogrammed complex soccer plays in appropriate situations by employing pl an-based control techniques. The use of such sophisticated state estimation and c o trol techniques characterizes the AGILO software. The paper discusses the c omputational techniques and necessary extensions based on experimental data from the 2001 robot soccer world championship.