Action utility forecasting based role implementation in robot soccer system

The robot soccer strategy system based on the dynamic role transition algorithm is highly recommended in recent years among robot soccer teams and research institutions. The object of the paper is to enhance the success rate and coordination performance of the role execution subsystem. The paper studied the role implementation mechanism of the robot soccer strategy system in the offensive situation based on the role-based decision-making system. First, the concept of maneuver area was proposed. Then, a method of achieving coordination among soccer robots through predicting the utility of the team after one step of role actions, and then choosing the role action combining the concepts and methods of game theory was presented. The results in the FIRA 2D simulation platform show that the proposed method performs well in selecting role behavior, and is efficient in helping the team to achieve good teamwork.