面向多状态路网的交通控制子区动态划分方法研究

项目名称： 面向多状态路网的交通控制子区动态划分方法研究

项目编号： No.61304198

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 自动化技术、计算机技术

项目作者： 别一鸣

作者单位： 哈尔滨工业大学

项目金额： 23万元

中文摘要： 交通控制子区动态划分是自适应交通控制系统的关键技术之一，它对于提高路网交通控制效益、缓解交通拥堵具有重要意义。受控的城市路网一般包含多种状态级别的信号交叉口，不同状态的交叉口需要不同的控制目标与控制策略，进而需要不同的子区划分方法。基于此，本项目以交通控制子区动态划分方法为研究对象，考虑受控路网交叉口状态的多样性，建立交叉口状态划分方法，提出交通控制子区多目标划分策略，生成与划分目标对应的关联度指标集；以子区划分效益函数为基础，量化路网动静态交通要素对关联度指标的影响机理，构建各交通状态下关联度指标的量化方法、多影响因素联合作用下的相同状态等级相邻交叉口、不同状态等级相邻交叉口整体关联度计算模型；研究初始路网状态下的子区划分算法，优化子区划分方案动态调整启动阈值，建立划分方案动态调整算法。研究成果可为复杂路网条件下交通控制子区的科学划分提供理论依据。

中文关键词： 交通控制子区；关联度模型；划分算法；信号控制；多状态

英文摘要： Dynamic partition of traffic control subarea is one of the critical techniques of adpative traffic control system. It is meaningful to improve the traffic control benefits and alleviate traffic congestion. Generally, the signalized intersections in the controlled ubran road network have multiple traffic states. The intersections in different states need different control objectives and strategies, and then need different subarea partition methods. Therefore, this project takes the dynamic partition method of traffic control subarea as research object. By considering the diversity of intersection states, the states partition algorithm is brought forward and then the subarea partition strategy with multiple objectives is developed. Corresponding to the multiple objectives, the correlation index set is generated. Based on the subarea partition benefits function, the impact mechanism of dynamic and static factors of road network on multiple correlation indices are quantified, and the methods to quantify the correlation indices in multiple traffic states are established. For the adjacent intersections with the same traffic states and different traffic states, the correlation models are developed under the joint impacts of multiple factors respectively. The subarea partition algorithm for original netwrok is studied,

英文关键词： Traffic control subarea；correlation model；partition algorithm；signal control；multiple traffic states