交通出行诱导的Stackelberg博弈模型及其遗传算法求解  

Stackelberg Game Model and Calculating Method by GA for Traffic Guidance Strategy

作　　者：李文勇[1,2] 陈学武[1] 陈峻[1] 

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

出　　处：《武汉理工大学学报（交通科学与工程版）》2009年第5期984-987,共4页Journal of Wuhan University of Technology（Transportation Science & Engineering）

基　　金：国家自然科学青年基金项目(批准号:50808050);国家"十一五"科技支撑计划项目(批准号:2006BAJ18B01);广西科学研究与技术开发计划项目(批准号:桂科攻0719001-2)资助

摘　　要：针对交通出行诱导的实际需求,分析了进行博弈的共同知识.基于对道路交通管理者和道路使用者的策略分析,通过建立道路使用者确信度模糊模型,构建了交通出行诱导的离散动态Stackelberg博弈模型.针对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.设计了基于遗传算法的博弈模型计算机求解算法,使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.Aiming at actual traffic travel guidance demand, the common information for gaming was analysed. The strategies of traffic management and road users were studied, and a differential Stackelberg game model which could describe the guidance strategies was constructed based on the established fuzzy model of road users. After two phases game character of traffic travel guidance were summarized, converse induction method to solve game process was brought out. The calculating method by Genetic Algorithm （GA） to solve Stackelberg game model was devised. And the travel guidance scheme solved would achieve the aim of the individual user optimum based on system optimum. The analysis result for the example road-net shows that the game model and calculating method brought out in the paper would importantly conduce to produce the traffic guidance scheme for traffic guidance system from the aspect of theory and technology.

关 键 词：出行诱导 博弈模型 遗传算法 

分 类 号：U491[交通运输工程—交通运输规划与管理] O225[交通运输工程—道路与铁道工程]