【作者】 李力；

【导师】 李龙江；

【作者基本信息】 电子科技大学 ， 通信与信息系统， 2014， 硕士

【摘要】 随着社会的高速发展,交通拥堵问题严重的影响了国家城市化的进程。智能交通系统作为一门新兴的研究课题,为解决城市交通拥堵问题提供了新的解决方案。交通控制和车辆诱导作为智能交通系统的两个重要组成部分,在解决城市交通拥堵问题上越来越受到关注和重视。在现有的智能交通系统研究中,主要利用交通控制子系统对路况进行处理,而车辆诱导子系统完成对车辆的路径规划,大多偏重于对交通灯控制和车辆诱导的分开研究。但是在实际的城市交通路网中,通过将交通灯控制与车辆诱导进行集成,能够进一步促进道路拥堵的疏散,且有助于提高出行者的行车效率。本文针对交通灯控制与车辆诱导的集成进行了研究,针对车辆诱导的具体实现场景,提出了基于交通灯配置的车辆诱导机制,该车辆诱导机制主要通过两个步骤来完成对路网中道路拥堵的消散以及车辆的诱导调度,这两个步骤分别是:(1)路网流量均衡策略:根据交通的动态路况自动配置交通灯的绿信比,完成对拥堵道路的疏散;(2)路网流量优化分配策略:对全网车辆进行诱导分配,达到整个路网中车辆总行驶时间最少。通过MATLAB仿真工具,对基于交通灯配置的车辆诱导机制进行了仿真,并比较分析了车辆诱导前后路网中平均饱和度、饱和度方差、全网车辆行驶时间等参数变化情况,实验结果表明,通过基于交通灯配置的车辆诱导机制能够有效地降低路网中的平均饱和度及饱和度方差,均衡整个路网的交通流量,同时能够降低全网车辆的总行驶时间,提高了路网的通行能力。此外,本文将交通灯调整时间作为车辆诱导算法中的一个时间因子,进一步提出了基于交通灯配置的车辆诱导算法ED*算法,通过车辆ED*诱导算法完成对路网中的车辆个体进行诱导,完成对其出行路径的动态规划,以达到使微观车辆个体行驶时间最少的目的。通过VC++6.0完成了对ED*车辆诱导算法的计算及仿真,并与A*算法、D*Lite算法进行比较,实验结果表明,在路况信息发生变化后,ED*算法在重新规划路径的计算速度上具有显著的提高,有助于提高路网中车辆的行驶效率,适用于城市动态交通路网。更多还原

【Abstract】 With the rapid development of society, the traffic congestion problem seriously affected the country’s urbanization process. Intelligent transportation systems as an emerging research topic provide a new solution to solve the problem of urban traffic congestion. Traffic control and vehicle-induced, as two important components of intelligent transportation systems, attract growing concern and attention in solving the urban traffic congestion problem. In current study on intelligent transportation systems, we usually use the road traffic control subsystem to solve the traffic problem, and the vehicle-induced subsystem to complete vehicular route planning. While most of the current research separately study on the control of traffic lights and vehicle-induced. However, in real urban traffic network, by integrating traffic lights control and vehicle-induced can further promote the evacuation of road congestion and help improve the efficiency of the traveler’s journey.In this paper, we mainly study the integration of traffic lights control and vehicle-induced. This paper proposes a vehicle-induced mechanism based on traffic lights configuration, and this mechanism can dissipate road congestion and complete vehicle scheduling by two steps. These two steps are:(1) traffic load balancing strategy: automatic configuration traffic lights green ratio under dynamic road traffic to dissipate road congestion;(2) traffic load optimal allocation strategy: induction vehicles on the road to minimize the total travel time. MATLAB simulation tool is used to simulate vehicle-induced mechanism based on traffic lights configurations. Compare and analyze the average saturation, saturation variance, and the whole network parameters such as vehicle time changes of the vehicle before and after induction of the road network. The experimental results show that this mechanism can effectively reduce the average saturation and saturation variance, balance whole road network traffic, reduce the total travel time of the whole network of the vehicle and improve the traffic network the capacity as well.In addition, this article treats the traffic lights configuration time as a factor to complete Enhanced D-star Lite algorithm(ED*), a vehicle-induced algorithm based on vehicle traffic lights configuration. Use ED* algorithm to complete single vehicle-induced and dynamic route program, in order to minimize individual vehicle travel time in microscopic view. VC++ 6.0 is used to simulate the vehicle-induced ED* algorithm, and compared with the A* algorithm and the D* Lite algorithm. The experimental results show that after traffic information changes, our ED* algorithm can significantly improve the speed of the path re-planning, help to reduce the vehicle travel time. This mechanism is suitable for urban dynamic road network.更多还原