路网拥塞控制中的多目标路径决策模型研究

发布时间：2018-04-16 12:01

本文选题：多目标优化 + 路网拥塞　；参考：《湖南大学学报(自然科学版)》2015年04期

【摘要】：智能交通系统领域中的路网拥塞控制是解决路网拥塞问题的主要手段之一,针对该问题,利用自底向上的agent建模方式,构建一种多目标路径决策agent移动模型.在该模型中,车辆agent兼顾最短路径和拥塞避免两个优化目标,通过车辆agent行驶距离最短(最短路径)和途经区域的拥塞程度最低(拥塞避免)两个目标优化来动态进行路径决策.基于多目标路径决策移动模型一方面能够实现对交通拥堵路段的分流控制,另一方面能够挖掘网络拓扑结构中易发生拥塞的路口的共同特征,为路网拥塞控制提供帮助.仿真实验结果表明,该模型能较好地改善路网结构中的拥塞路段.针对不同链路密度及链路分布的网络所进行的仿真实验结果进一步表明,路网结构的链路密度对拥塞路段出现在网络中的地理位置影响不同,而路口节点位置影响其拥塞程度;网络结构的链路分布形态对发生拥塞路段的地理位置和拥塞优化结果具有直接影响.
[Abstract]:Road network congestion control in the field of intelligent transportation system is one of the main methods to solve the problem of road network congestion. Aiming at this problem, a multi-objective path decision agent mobile model is constructed by using bottom-up agent modeling method.In this model, vehicle agent takes into account two optimization objectives: shortest path and congestion avoidance.Dynamic path decision is made by optimizing the vehicle agent with the shortest distance (shortest path) and the least congestion degree (congestion avoidance) in the passing area.Based on the multi-objective path decision moving model, on the one hand, it can realize the traffic congestion control, on the other hand, it can mine the common characteristics of the congested intersection in the network topology, and provide the help for the road network congestion control.The simulation results show that the model can improve the congested section of road network structure.The simulation results for networks with different link density and link distribution further show that the link density of road network structure has different effects on the geographical location of congested sections in the network, and the location of intersection nodes affects the congestion degree.The distribution pattern of network structure has a direct impact on the location and congestion optimization of congested sections.
【作者单位】：湖南大学信息科学与工程学院;东京工业大学;
【基金】：湖南省自然科学基金资助项目(13JJ3049) 湖南省科技计划资助项目(2012FJ4131) 国家863高技术资助项目(2012AA01A301-01)
【分类号】：U495

【参考文献】