The invention discloses a traffic state estimation method based on fixed and mobile traffic monitoring data, which is different from a previous multi-source data fusion method for traffic state estimation, and introduces data fusion of a BP neural network algorithm based on a genetic algorithm. And fusing the travel time data obtained by the fixed detector data and the floating car GPS data. The improved neural network algorithm shortens the training time, ensures the optimal solution of data fusion, improves the efficiency of data fusion, and the fusion data obtained after fusion processing reflects more real and reliable traffic information, can be effectively used for traffic state estimation, and improves the accuracy of traffic state estimation.
本发明公开了一种基于固定和移动交通监测数据的交通状态估计方法,不同于以往的交通状态估计的多源数据融合方法,引入了基于遗传算法的BP神经网络算法的数据融合,对由固定检测器数据和浮动车GPS数据得到的行程时间数据进行融合。改进神经网络算法缩短了训练耗时,保证了数据融合的最优解,提升了数据融合的效率,并且经过融合处理后所获取融合数据反映了更加真实可靠的交通信息,能有效用于交通状态估计,提高交通状态估计的准确度。
Traffic state estimation method based on fixed and mobile traffic monitoring data
一种基于固定和移动交通监测数据的交通状态估计方法
10.03.2023
Patent
Elektronische Ressource
Chinesisch
Traffic state estimation method only based on mobile traffic detection data
Europäisches Patentamt | 2024
|Macroscopic Traffic State Estimation: Understanding Traffic Sensing Data-Based Estimation Errors
DOAJ | 2017
|