摘要
针对带有时间约束的、可以动态加入到环境中的复杂任务,建立了一种基于对策论的任务协作模型,该模型至少存在一个纯策略Nash平衡解.给出了一种任务协作方法,该方法中Agent能够根据自身的局部信息进行行为选择,并利用虚拟行动学习方法确保Agent快速学习到一个纯策略Nash平衡,仿真实验结果表明该方法与Chapman和刘的方法同等有效.
For the complex tasks with time constraints,which can dynamically be added to environment,a task coordination model based on game theory has been established,which has at least one pure Nash equilibrium.A task cooperation method has been proposed,which makes agent choose its behavior according to the local information and ensure that agent learns a pure strategy Nash Equilibrium quickly by using fictitious play learning method.Simulation results show that this method is as effective as Chapman and Liu＇s methods.
出处
《河南师范大学学报（自然科学版）》
CAS
北大核心
2013年第4期158-161,共4页
Journal of Henan Normal University(Natural Science Edition)
基金
河南省重点科技攻关项目(102102210179
102102210176
122102210086)
河南省教育厅自然基金项目(13A520530)
