| 基于NashCC-Q学习的两交叉口信号灯协调控制 |
| |
| 引用本文： | 赵晓华,李振龙,于泉,李云驰.基于NashCC-Q学习的两交叉口信号灯协调控制[J].系统仿真学报,2008,20(17). |
| |
| 作者姓名： | 赵晓华 李振龙 于泉 李云驰 |
| |
| 作者单位： | 1. 北京市交通工程重点实验室,北京工业大学,北京,100124
2. 北京工业大学电子信息与控制工程学院,北京,100124 |
| |
| 基金项目： | 国家自然科学基金，北京市属市管高等学校人才强教计划 |
| |
| 摘 要： | 提出一种NashCC-Q学习算法用于解决两交叉口信号灯协调控制问题.根据博弈论概念,相邻两交叉12之间的协调控制问题属于二人非零和合作博弈类型.在Nash.Q学习算法的基础上,将Q值函数的更新建立在Nash合作博弃中提出的Nash"公理方法"求博弈谈判解的基础上,进而解决合作博弈的问题,实现两交叉口信号灯协调控制.采用Paramics仿真软件进行仿真,结果表明该方法的有效性.
|
| 关 键 词： | Q学习算法 博弈论 Nash公理谈判解 两交叉口信号灯协调控制 |
Traffic Signal Coordination Control for Two Adjacent Intersections Based on NashCC-Q Learning |
| |
| ZHAO Xiao-hua,LI Zhen-long,YU Quan,LI Yun-chi.Traffic Signal Coordination Control for Two Adjacent Intersections Based on NashCC-Q Learning[J].Journal of System Simulation,2008,20(17). |
| |
| Authors: | ZHAO Xiao-hua LI Zhen-long YU Quan LI Yun-chi |
| |
| Abstract: | The NashCC-Q learning algorithm was proposed to resolve traffic signal coordination control for two adjacent intersections. The problem of coordination control for two adjacent intersections is the type of two-player general sum cooperative game in Game theory. Therefore,according to the Nash-Q learning algorithm,the update of the Q-Value function was based on Nash bargaining solution,furthermore,coordination control for two adjacent intersections was realized. The validity of this method was proved by Paramics simulation software. |
| |
| Keywords: | Q-Learning algorithm game theory Nash Bargaining Solution trraffic signal coordination control for two adjacent intersections |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
