基于车路协同的单点信号交叉口自适应优化控制方法.docVIP

基于车路协同的单点信号交叉口自适应优化控制方法 周建山 田大新 韩 旭 王云鹏 （北京航空航天大学 交通科学与工程学院， 北京 100191） 摘 要：首先建立了面向城市单点信号交叉口的信息交互网络的逻辑架构，定义了基于车路协同的交通流数据采集、传输的原型场景；在此基础上，将神经计算原理与模糊决策技术引入信号优化控制策略的设计中，分别建立起基于极大熵原理的交通OD矩阵估计模型、SOM神经网络模型以及Takagi-Sugeno-Kang模糊决策系统，对单点信号交叉口的交通流模式进行有效识别和优化控制。在Visual Studio 2010软件平台上，基于Visual Basic.Net语言对VISSIM微观交通仿真软件进行了二次开发，搭建起用于仿真测试的单点信号交叉口模型，并在MATLAB平台上编制控制策略的算法程序；基于接口编程技术，利用Visual Studio平台进行VISSIM与MATLAB计算引擎协同仿真，测试了本文设计的控制策略以及传统的定时控制策略和非自适应动态控制策略。通过对比分析交叉口在不同控制策略下的效能指标，表明本文设计的控制策略具有较高的信号优化性能。 关键词：智能交通；车路协同系统；信号控制；神经计算原理；模糊逻辑 A Self-adaptive Optimal Control Approach for Urban Isolated Signalized Intersection Based on VII Jianshan Zhou，Daxin Tian，Xu Han，Yunpeng Wang (School of Transportation Science and Engineering, Beihang University, Beijing 100191, China) Abstract: This paper presents the logic framework of a traffic information communication network utilized in an urban isolated signalized intersection, which can be made available from VII system. Based on the VII system, a prototype scenario for collecting and transmitting traffic flow data is then defined in details, and is used for the traffic control modeling and the related algorithms designing. Moreover, the paper introduces neural computing principle and fuzzy decision making technique into the control strategy for optimizing the signal. By using Visual Basic.Net language, the collaborative simulation platform based on the integration of VISSIM and MATLAB engine is built to verify our control strategy. What’s more, two traditional signal control strategies: fixed time control and non-adaptive dynamic control are also tested on the simulation platform. By comparing the performances of these control strategies, we can 基金项目：国家自然科学基金资助项目 draw the significant conclusion that the proposed control strategy has the overwhelming advantages due to its self-adaptive ability and optimization capability. Key words: Intelligent Transportation System, Vehicle-Infrastruc