计算机技术与发展2025，Vol.35Issue(11)：12-19,8.DOI:10.20165/j.cnki.ISSN1673-629X.2025.0151
公路交通流预测综述:方法与进展
A Survey on Highway Traffic Flow Prediction:Methods and Advances
摘要
Abstract
With the acceleration of urbanization and the rapid development of intelligent transportation,traffic flow prediction has become a core research direction in the fields of smart transportation and smart cities.We systematically review the research status of traffic flow prediction technologies based on deep learning,and thoroughly analyze the technical challenges faced in current traffic flow prediction,in-cluding insufficient modeling of spatiotemporal dependencies,the complexity of external factors,data quality issues,as well as the high computational cost and poor interpretability of deep learning models.At the same time,we point out that future research should focus on the collaborative optimization of"data-model-system"in three major areas:at the data layer,efforts should be made to enhance multimodal data fusion and privacy protection;at the model layer,the integration of knowledge graphs and large models can improve the dynamic spatiotemporal modeling capability;at the system layer,edge intelligence and distributed computing should be leveraged to improve real-time performance and system robustness.The research provides important theoretical support and practical guidance for the further development of intelligent transportation systems.关键词
交通流预测/时空依赖建模/深度学习/多模态数据/智能交通Key words
traffic flow prediction/spatiotemporal dependency modeling/deep learning/multimodal data/intelligent transportation分类
计算机与自动化引用本文复制引用
王小博,张轩,高张浩,董云卫..公路交通流预测综述:方法与进展[J].计算机技术与发展,2025,35(11):12-19,8.基金项目
陕西省交通运输2023年度交通科研项目(23-107K) （23-107K）
