不同交通流状况下车辆分布式协同调度仿真

Distributed Cooperative Vehicle Scheduling Simulation under Different Traffic Flow Conditions

面对区域内不同交通流状况,进行有效的车辆调度对于缓解城市交通压力、提高道路资源利用率具有重要的现实意义。针对定时控制、感应控制和自适应控制三种车辆调度方法性能不足,导致交通拥堵指数较大的问题,研究一种车辆分布式协同调度方法。方法研究分为三部分:首先利用BP神经网络算法判别交通流状况,然后分别分析交通信号控制与交通诱导系统的工作原理,最后构建交通信号控制与交通诱导协同调度模型,实现车辆均衡分流。结果表明:与定时控制、感应控制和自适应控制三种传统调度方法相比,所提方法应用后,交通拥堵指数降低,从车辆平均延误时间、车辆平均停车次数、平均行程速度、单位时间内车流量、平均排队长度等几项指标上得以体现,由此说明本方法更能缓解交通拥堵情况,调度性能更好。

In the face of different traffic flow conditions within each region,effective vehicle scheduling is significant for alleviating urban traffic pressure and improving the utilization ratio of road resource.Therefore,a distributed cooperative vehicle scheduling method was studied.This method was divided into three parts.Firstly,the BP neural network algorithm was used to identify traffic flow conditions,and then the working principle of traffic signal control and the traffic guidance system were analyzed respectively.Finally,the traffic signal control model and traffic guidance cooperative scheduling model were constructed to divide the vehicle flow.Simulation results show that,compared with traditional methods such as timing control,induction control and adaptive control,the proposed method decreases traffic congestion indexes.These can be reflected in the average delay time,average parking times,average speed,vehicle flow per unit time and average queue length.This method can better alleviate traffic congestion.In this case,the scheduling performance is better.