基于Q学习的Agent智能防守策略研究与应用

Research and Application about Defensive Strategy Based on Q Learning

模拟机器人足球比赛(Robot World Cup,RoboCup)作为多Agent系统的一个通用的实验平台,通过它可以来评价各种理论、算法和框架等,已经成为人工智能的研究热点。针对RoboCup仿真中的守门员防守问题,基于Q学习算法,描述了在特定场景中应用Q学习训练守门员的方法和过程。在RobCup中验证了该算法,实现了守门员防守策略的优化。

As a representative experimental platform of multiagent system, RoboCup（Robot World Cup） by which various theories, algorithms and architectures can be evaluated, has become the research center of artificial intelligence. To rcsolve the problem about defensive strategy of goalie in RoboCup environment, based on 0 learning proposed a method which trained goalie. Confirm the algorithm in RoboCup environment and implement the optimization of defensive strategy about goalie.