The invention provides a traffic flow speed prediction method based on a heterogeneous information network and a space-time attention mechanism, and the method comprises the steps: 1, carrying out the data collection of historical traffic flow data collected at an intersection, and obtaining a training sample; step 2, constructing a heterogeneous information network training model by acquiring microscopic nodes of the road microscopic graph and regional nodes of the regional node graph; step 3, traffic flow prediction training is carried out on historical traffic flow data in the training sample through the heterogeneous information network training model; and step 4, through integration of heterogeneous information of different stages, obtaining spatial-temporal characteristics of the microscopic nodes and the regional nodes in different stages, and carrying out fusion and extraction so as to predict the traffic flow speed and obtain a prediction result.
本发明提供了一种交通流速度的预测方法,基于异质信息网络与时空注意力机制,包括:步骤1,对在路口处收取到的历史交通流数据进行数据采集,获取训练样本;步骤2,通过获取道路微观图的微观节点和区域节点图的区域节点构造异质信息网络训练模型;步骤3,通过所述异质信息网络训练模型对所述训练样本中的历史交通流数据进行交通流预测训练;步骤4,通过整合不同阶段的异质信息,获取所述微观节点和所述区域节点不同阶段的时空特征进行融合和提取,用于对交通流速度进行预测,得到预测结果。
Method and device for predicting traffic flow speed
一种交通流速度的预测方法及装置
2022-09-06
Patent
Electronic Resource
Chinese
Apparatus for predicting traffic speed and method thereof
European Patent Office | 2024
|