系统仿真学报 ›› 2017, Vol. 29 ›› Issue (11): 2782-2787.doi: 10.16182/j.issn1004731x.joss.201711026

• 仿真系统与技术 • 上一篇    下一篇

Robocup2D项目中Agent2D底层动作链机制的分析优化

陈冰1, 许非凡2, 徐涵延2, 程泽凯3, 刘诚1   

  1. 1.信息工程大学理学院,河南 郑州 450001;
    2.信息工程大学指挥军官基础教育学院,河南 郑州 450001;
    3.安徽工业大学计算机学院,安徽 马鞍山 243002
  • 收稿日期:2016-06-24 发布日期:2020-06-05
  • 作者简介:陈冰(1982-),男,湖北孝感,硕士,讲师,研究方向为系统分析与集成;许非凡(1993-),男,北京,本科,研究方向为仿真系统及实现;徐涵延(1994-),男,贵州,本科,研究方向为仿真系统及实现。

Analysis and Optimization of the Action Chain Mechanism in Agent2D Underlying in RoboCup2D Soccer League

Chen Bing1, Xu Feifan2, Xu Hanyan2, Cheng Zekai3, Liu Cheng1   

  1. 1. School of Arts and Sciences, Information Engineering University, ZhengZhou City, 450001, China;
    2. School of Junior Commanding Officers, Information Engineering University, ZhengZhou City, 450001, China;
    3. School of Computer Science, AnHui University of Technology, Ma Anshan City, 243002, China
  • Received:2016-06-24 Published:2020-06-05

摘要: 在RoboCup2D仿真足球项目中,Agent2D是我国使用最为广泛的球队底层之一。仿真平台中数据传输的噪声干扰及代码自身动作链机制不完整等因素,导致采用Agent2D底层的球队在应对不同的队伍时,存在着适应能力不足的缺点,影响了球队的整体能力。该论文引入了动作修正参数,利用强化学习的手段对动作链机制进行优化,使Agent底层球队在面对不同风格的对手时可以选择更加有效的动作执行,以此来提升球队的适应性。仿真实验证明,此法具有一定效果。

关键词: RoboCup2D, 仿真, 动作链, 动作修正参数, 强化学习

Abstract: In the RoboCup2D soccer league, Agent2D is one of the most widely used underlying team in China. Data transmission noise and the incomplete action chain mechanism make the underlying teams using Agent2D be lack of flexibility. This paper introduces an action correcting parameter and optimizes the operation of the action chain by reinforcement learning mechanism. The performance of the Agent2D underlying team is improved in the game and the adaptability of the team is enhanced. Simulation experiment results show that this method has a certain effect.

Key words: RoboCup2D, Simulation, action chain, action correcting parameter, reinforcement learning

中图分类号: