The invention discloses a traffic speed prediction method based on a multi-spatial scale space-time Transform, and belongs to the technical field of traffic prediction planning. The prediction method comprises the following steps: sequentially inputting preprocessed road section sensor speed sequence data into a multi-scale spatial feature extraction module, a traffic spatial-temporal feature extraction module and a prediction module; feature extraction of a multi-scale dynamic space structure and a static road network structure is gradually realized, space-time dependence is accurately modeled, and the traffic speed in a period of time in the future is predicted. The multi-scale spatial feature extraction module can comprehensively and specifically extract spatial features, and a large amount of useless calculation is reduced while the prediction precision is improved. In addition, the traffic spatial-temporal feature extraction module selects more valuable historical data according to the traffic characteristics and the relative position information of the data so as to perform sufficient spatial-temporal feature extraction, and the problem that the relative position information is lost when spatial-temporal dependence is extracted is solved.
本发明公开了一种基于多空间尺度时空Transformer的交通速度预测方法,属于交通预测规划技术领域。预测方法包括:将预处理后的路段传感器速度序列数据依次输入多尺度空间特征提取模块、交通时空特征提取模块以及预测模块,逐步实现多尺度动态空间结构和静态路网结构的特征提取、精准建模时空依赖以及预测未来一段时间的交通速度。本发明多尺度空间特征提取模块能够全面且有针对的提取空间特征,在提高预测精度的同时减少了大量无用计算。另外,交通时空特征提取模块根据交通特性以及数据的相对位置信息选择更有价值的历史数据以进行充分的时空特征提取,解决了提取时空依赖时存在的丢失相对位置信息的问题。
Traffic speed prediction method based on multi-spatial scale space-time Transform
一种基于多空间尺度时空Transformer的交通速度预测方法
2023-06-23
Patent
Elektronische Ressource
Chinesisch
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Traffic flow prediction method based on improved space-time Transform
Europäisches Patentamt | 2022
|Time-space traffic state prediction method based on Transform network
Europäisches Patentamt | 2021
|DSpace@MIT | 2014
|Traffic speed prediction method based on multi-space-time diagram fusion convolutional network
Europäisches Patentamt | 2023
|Europäisches Patentamt | 2023
|