无信号交叉口车辆调度方法研究

Research on Vehicle Scheduling Optimization of Unsignalized Intersection

随着车辆数量的快速增加,交通拥堵问题变得日益严重.如何引导车辆安全高效地通过交叉路口已经受到了学界的广泛关注.已有的控制方法主要是在线优化信号灯的相位配比.然而相位之间的频繁切换会导致信号周期中黄灯时间占比的增加,进而降低交叉口的车辆放行能力.本文提出了一种基于车路协同的无信号交叉口资源调度模型,该模型将交叉口划分为互不相交的物理空间路权资源,并描述了各个路权资源之间的相互协同关系,进而将无信号交叉口交通控制问题转换为有限资源调度问题.在此基础上,构建最大化交叉口通行效率的目标函数,并求解车辆的最优通行序列.实验结果表明:较传统有信号交叉口控制方法,无信号控制方法有效减少了车辆的排队长度,提高了交叉口的车辆吞吐能力.

With the rapid growth of the number of vehicles,the problem of traffic congestion has become increasingly serious.How to guide a vehicle to pass the intersection safely and efficiently has been widely concerned by the academic community.The existing control methods mainly focus on optimizing the signal phase on-the-fly.However,frequent switching of the phases will lead to an increase in the ratio of yellow light time in a signal cycle,which decreases the traffic flowing capacity of an intersection.In this paper,a microscopic model for unsignalized traffic intersection using automaton is proposed.The intersection is divided into different spatial traffic resources,and the model describes the relations of these spatial traffic resources.As the result,the traffic control problem at the unsignalized intersection is transformed into a scheduling problem with limited resource constraints.With this model,an objective function for improving the traffic condition of the intersection is established,and an optimal vehicle scheduling policy is calculated.The experimental results shows: compared with the corresponding results obtained using traditional signal control method,the scheme introduced in this paper can effectively reduce the length of the queued vehicles at the intersection,as the maximum throughput of that intersection is obviously increased.