Multi-agent strategic modeling in a robotic soccer domain
Andraž Bežek, Matjaž Gams, Ivan Bratko · 2006
This paper presents an algorithm for multi-agent strategic modeling (MASM) applied in a robotic soccer domain. It transforms a multi-agent action sequence into a set of strategic action descriptions in a graphical and symbolic form. By using hierarchically ordered domain knowledge, the algorithm is able to generate graphic and symbolic strategic action descriptions together with corresponding rules at different levels of abstraction. The method was evaluated on the RoboCup Soccer Server Internet League data.