Using a Datawarehouse to Extract Knowledge from Robocup Teams

RoboCup is a scientific and educational, international project that involves artificial intelligence, robotics and sport sciences. In these competitions, teams of all around the world participated in distinct leagues. In the begining of the Coach competition, one of RocoCup leagues,the goal of the researchers was to develop an agent (Coach that provides advices to teammates about how to act and with improve team performance. Using the resulting improved coach agent, with enhanced statistic calculation abilities, a huge amount of statistical data was gathered from the games held at Bremen 2006. This data was then stored and treated in a data warehouse system obtaining a good high level perspective/knowledge of the RoboCup simulated soccer tournament. According to the results, the team that represented our country, has a much more goal opportunities in comparison with the majority of the teams, but this team did not score many goals. In terms of more occupied regions, the best four teams in the tournament did not occupy many times the left and right wings, compared to others regions. In the future the our country team needs to develop new strategies that use these two areas preferrentialy in order to achieve better results.