引用本文:程泽凯,管博伦,郭志浩,秦峰.基于进攻关键区域的RoboCup2D球队防守策略的优化(J/M/D/N,J:杂志,M:书,D:论文,N:报纸).期刊名称,2021,38(3):26-35
CHEN X. Adap tive slidingmode contr ol for discrete2ti me multi2inputmulti2 out put systems[ J ]. Aut omatica, 2006, 42(6): 4272-435
【打印本页】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 1059次   下载 22 本文二维码信息
码上扫一扫!
分享到: 微信 更多
基于进攻关键区域的RoboCup2D球队防守策略的优化
程泽凯,管博伦,郭志浩,秦峰
安徽工业大学 计算机科学与技术学院,安徽 马鞍山 243002
摘要:
在RoboCup仿真2D足球比赛中,防守模型是关键技术之一,加强在对方进攻关键区域上的防守可以有效阻止对手进球,针对RoboCup仿真2D球队防守策略难以优化的问题提出一种基于球队进攻关键区域优化防守策略的方法;方法通过对比赛过程中的日志文件进行解析,使用数据挖掘方法提取动作链和传球热力图解释了球队进攻有侧重点的合理性,利用比赛过程中球的位置分布对球场进行了更合理的动态的离散划分,在每个离散区域中采用自适应参数的密度聚类算法来计算球队进攻关键区域,并且依据这些区域和进攻过程中的重点球员设计了防守策略;实验部分依据防守策略优化了底层代码,结果表明:对进攻关键区域进行重点防守能有效阻止对方的进攻,同时也验证了进攻关键区域计算的正确性。
关键词:  进攻关键区域  数据分布  日志文件  密度聚类
DOI:
分类号:
基金项目:
Optimization of Defensive Strategies of RoboCup2D Teams Based on Attacking Key Areas
CHENG Ze-kai, GUAN Bo-lun, GUO Zhi-hao, QIN Feng
School of Computer Science and Technology, Anhui University of Technology, Anhui Maanshan 243002, China
Abstract:
In RoboCup simulated 2D football match, defensive model is one of key techniques, and strengthening the defense in the key area of the opponent's attack can effectively prevent the opponent from scoring. According to the problem in the difficulty in optimizing the defensive strategies for RoboCup simulated 2D teams, this paper proposes a method to optimize the defensive strategies based on the team's attack key areas. This method explains the rationality for the emphasis on some offenses by analyzing the log files in the process of the games and by using data mining method to extract motion chain and the thermodynamic chart of the passing, makes more reasonable dynamic discrete division for the field by ball position distribution in the process of the game, uses density clustering algorithm with adaptive parameters to calculate the key areas of the attack of the teams in every discrete area, and designs the defensive strategies based on the key players in these areas and in the process of the attacks. The underlying code is optimized according to the defensive strategy in experiment part. Results show that the key defense of the offensive key areas can effectively prevent the attack of the other side, and at the same time, the correctness of the calculation of the offensive key areas is verified.
Key words:  attack key area  data distribution  log file  density clustering
重庆工商大学学报(自然科学版) 版权所有
地址:中国 重庆市 南岸区学府大道19号 重庆工商大学学术期刊社 邮编:400067
电话:023-62769495 传真:
您是第4752840位访客
关注微信二维码