期刊文献+

基于影响图的传球训练策略的实现 被引量:1

Implementation of Strategy of Passing Based on Influence Diagrams
下载PDF
导出
摘要 影响图是决策问题的图形表示,利用影响图可以有效地解决智能Agent的行为选择,而传球问题就是球员Agent选择传球速度和角度的问题。建立了一个基于影响图的传球模型,并根据该模型创建一个评价传球效果的效用函数;然后利用由教练程序改写的记录员程序来记录数据,并进行统计得到效用函数中的概率值;最后通过训练得到效用函数中系数的值,并利用此效用函数作传球训练。训练结果表明该效用函数是可信的。 Influence diagram is a diagram of decision- making, using influence diagrams can solve the behavior choice of intelligence agent effectively, while the question of passing help the agent choose the speed and angle. In this paper, first build a model of passing and a utility function which can evaluate passing decision. Second,data are recorded by recorder program which is recomposed by coach program and then calculate these data to obtain the probability of utility function. In the end, the values of coefficients in utility function gain by training, and then use the utility function to train the team. The training result indicates that the utility function is credible.
作者 张润梅
出处 《计算机技术与发展》 2008年第12期245-247,250,共4页 Computer Technology and Development
基金 安徽省自然科学基金(2006KJ036B) 安徽建筑工业学院青年基金(2005jq1144)
关键词 机器人足球 影响图 训练 传球 robocup influence diagrams training passing
  • 相关文献

参考文献9

二级参考文献30

  • 1张颖霞,杨宜民,陈波,岑汉彬.多智能体团队合作在机器人足球赛中的应用[J].微机发展,2004,14(7):112-114. 被引量:8
  • 2于磊,王浩,王骋.RoboCup中传球策略研究[J].计算机工程与应用,2004,40(28):59-61. 被引量:9
  • 3汤哲 孙增圻.机器人足球中的协作[A]..第四届全球智能与自动化大会论文集[C].上海,2002,6.469~473.
  • 4陈荣,徐用懋,兰鸿森.多层前向网络的研究——遗传BP算法和结构优化策略[J].自动化学报,1997,23(1):43-49. 被引量:51
  • 5D Cannel Markovitch S. Learning models of intelligent agents [A].Proceedings of the Thirteenth National Conference on Artificial Intelligence and the Eighth Innovative Applications of Artificial Intelligence Conference[C].1996, Vol. 2: 62-67. http://citeseer.ni.nec.com/carme1961earning.html
  • 6Howard R A Matheson J E. Influence diagrams[J]. Howard R.A.Matheson J.E.(Eds.). Readings on the principles and applications ofdecision analysis, Strategic Decisions Group, Menlo Park, CA, 1984,11(2): 719-762.
  • 7Noh S Gmytrasiewicz P J. Agent modeling in antiair defense[A].Proceedings of the Sixth International Conference on User Modeling[C]. 1997.389-400.
  • 8J Pearl. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference[M]. Morgan Kaufmann, San Mateo, California CA, 1988.
  • 9Gerhard W. Multi-agent system S:A modem Approach to Distributed Artificial Intelligence[M]. Boston, MIT press, 1999.
  • 10D.Suryadi Piotr J.Gmytrasiewicz. Learning Models of Other Agents Using Influence Diagrams[EB/OL]. http://citeseer.nj.nec.com/suryadi991earning.html, 1999.

共引文献27

同被引文献3

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部