交通出行诱导的离散Stackelberg动态博弈模型及其求解算法  

Discrete dynamic Stackelberg game model for traffic guidance strategy and its solving method

作　　者：李文勇[1,2] 陈学武[1] 陆建[1] 

机构地区：[1]东南大学交通学院,江苏南京210096 [2]桂林电子科技大学机电工程学院,广西桂林541004

出　　处：《控制理论与应用》2009年第10期1157-1161,共5页Control Theory & Applications

基　　金：国家自然科学基金资助项目(50808050);"十一五"国家科技支撑计划资助项目(2006BAJ18B01);广西科学研究与技术开发计划资助项目(桂科攻0719001-2)

摘　　要：针对交通出行诱导的实际需要,分析了道路交通管理者和道路使用者的博弈策略和博弈特征.通过道路使用者信息的模糊模型,建立了交通出行诱导的离散动态Stackelberg博弈模型.分析了交通出行诱导的两阶段博弈特征,提出了基于逆向归纳法的博弈模型求解算法.通过对模型无约束转化,运用Monte-Carlo法对示例路网进行了计算机求解,求解结果表明产生的交通出行诱导方案能够实现系统最优下的用户最优.In order to develop the practical actual traffic guidance, we investigate the game strategies and characteristics between the traffic controllers and the road users. The discrete Stackelberg game model, which describes the guidance strategies, is developed based on the established fuzzy model of the road user＇s information. After analyzing the game characteristics in two phases of traffic guidance, we put forward a solving method for this game model based on the converse induction method. Through the unconstraint model-conversion, an illustrative road-net is solved in a computer by using the Monte-Carlo method. The obtained traffic guidance scheme is optimal to the individual user as well as the whole system.

关 键 词：出行诱导 博弈模型 Monte—Carlo法 

分 类 号：U491[交通运输工程—交通运输规划与管理]