ニューラルネットワークを用いたロボットサッカーの戦術分析

抄録

With the progress of technology, the realization of a symbiotic society with human beings and robots sharing the same environment has become an important subject. An example of this kind of systems is soccer game. Soccer is a multi-agent game that requires strategies by taking into account each member’s position and actions. The behavior of a player changes depending on the game situation, such as the score gap, the remaining time and the importance of the match. In this paper, a futsal game played by agents is observed and analyzed to find a mapping relationship. The strategy in the team is analyzed based on the state vector parameters using a Self-Organizing Map (SOM) and different behaviors are analyzed and compared to achieve, in the future, the best possible result cooperation in a real game.