基于代理的OD矩阵反推模型

发布时间：2018-03-22 12:14

本文选题：交通　切入点：OD矩阵　出处：《计算机应用与软件》2016年12期 　论文类型：期刊论文

【摘要】：OD(Origin-Destination)矩阵是分析交通流、了解城市交通流特性并进行交通规划、管理、控制和诱导的重要基础和前提条件。基于代理有效反推OD矩阵,其依据道路外在环境和驾驶员认知、选择等内在因素对OD分布的影响,完成驾驶员对交通小区和行驶路线的决策,提出包括感知层、代理层、决策层和输出层四个层次的OD反推模型(ODABM)。实验中ODABM模型各时间段仿真车流量与路网实测车流量的平均范数平均相对误差减小到10.2%。结果表明:ODABM模型精度更高,更符合现实路网,并能反映交通流的时变性和出行目的分布,是提高交通仿真真实度的有效方法。
[Abstract]:ODN Origin-DestinationMatrix is an important foundation and prerequisite for analyzing traffic flow, understanding urban traffic flow characteristics, and planning, managing, controlling and guiding traffic. Based on the agent efficient backstepping of OD matrix, it is based on the external environment of the road and driver's cognition. The influence of selection and other internal factors on OD distribution, complete the driver's decision on the traffic area and route, and put forward the perception layer, agent layer, etc. In the experiment, the average relative error between the average norm of the simulated traffic flow and the measured traffic flow of the ODABM model is reduced to 10.2. The results show that the accuracy of the one-to-one ODABM model is higher, and it is more in line with the reality of the road network. It can reflect the time variation of traffic flow and travel purpose distribution, which is an effective method to improve the real degree of traffic simulation.
【作者单位】：南京邮电大学通信与信息工程学院;
【分类号】：U491

【相似文献】