Forecasting traffic flow is vital for the efficient operation of modern transportation systems and plays a pivotal role in advanced intelligent traffic management. Enhancing forecasting accuracy hinges on precisely grasping dynamic changes and spatial-temporal relationships of traffic flow. This study mainly focuses on such patterns as trends, periodicity, and spatial heterogeneity and proposes a time-delay aggregation transformer model (short for TDATF) to forecast traffic flow. The model leverages a time-delay aggregation and a self-attention mechanism to extract the trends and periodicities in traffic flow effectively, and a dynamic graph convolution module dynamically captures spatial dependency relationships. Meanwhile, a spatial position embedding matrix is utilized, which is constructed by a Laplacian-smoothed embedding vector derived from the Graph Convolutional Network (GCN) layer, to effectively model spatial heterogeneity. Experimental results on datasets PEMS03/04/07/08 demonstrate that the proposed TDATF model outperforms the baseline models in terms of forecasting accuracy.
ZhaoYLinYWenH, et al. Spatial-temporal position-aware graph convolution networks for traffic flow forecasting. IEEE Trans Intell Transp Syst2023; 24(8): 8650–8666.
2.
Medina-SalgadoBSánchez-DelacruzEPozos-ParraP, et al. Urban traffic flow prediction techniques: a review. Sustain Comput Inf Syst2022; 35: 100739.
3.
RahmaniSBaghbaniABouguilaN, et al. Graph neural networks for intelligent transportation systems: a survey. IEEE Trans Intell Transp Syst2023; 24(8): 8846–8885.
4.
TangGZLiBDaiHN, et al. SPRNN: a spatial-temporal recurrent neural network for crowd flow prediction. Inf Sci2022; 614: 19–34.
5.
ParkBJungB.Speed prediction of urban freeway using LSTM and CNN-LSTM neural network. J Korea Inst Intell Transp Syst2021; 20(1): 86–99.
6.
WuYZhaoLYuanZ, et al. CNN-GRU ship traffic flow prediction model based on attention mechanism. J Dalian Marit Univ2023; 49(1): 75–84.
7.
SongJFChenY.Traffic flow state prediction based on space-time correlation of vehicle trajectory using the deep hybrid model. J Intell Fuzzy Syst2023; 44(4): 6855–6863.
8.
ZhangHYangGYuH, et al. Kalman filter-based CNN-BiLSTM-ATT model for traffic flow prediction. Comput Mater Continua2023; 76(1): 1047–1063.
9.
HuangXTMaCXZhaoY, et al. A hybrid model of neural network with VMD-CNN-GRU for traffic flow prediction. Int J Mod Phys C2023; 34(12): 2350159.
10.
RenCXChaiCXYinC, et al. Short-term traffic flow prediction: a method of combined deep learnings. J Adv Transp2021; 2021: 15.
11.
LiJLiuYZouL. Dynamic graph convolutional neural network based on spatiotemporal modeling. J Peking University (Natural Science Edition), 2021, 57(04): 605–613.
12.
YuBYinHTZhuZ. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. In: 27th international joint conference on artificial intelligence (IJCAI), Stockholm, Sweden, 13–19 July 2018, pp: 3634–3640. Freibug: IJCAI.
WuZHPanSRLongG, et al. Graph WaveNet for deep spatial-temporal graph modeling. In: 28th international joint conference on artificial intelligence, Macao, China, 10–16 August 2019, pp:1907–1913. Freibug: IJCAI.
15.
FangZLongQSongG, et al. Spatial-temporal graph ODE networks for traffic flow forecasting. In: Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, Singapore, 14–18 August 2021, pp.364–373. New York: ACM.
16.
ZhangQChangWLiC, et al. Attention-based spatial-temporal graph transformer for traffic flow forecasting. Neural Comput Appl2023; 35(29): 21827–21839.
17.
YangSZhouYWuZ.Traffic flow prediction with random walks on graph and spatiotemporal bidirectional attention transformer. Appl Sci2024; 14(11): 4481.
18.
BoquetGMorellASerranoJ, et al. A variational autoencoder solution for road traffic forecasting systems: missing data imputation, dimension reduction, model selection and anomaly detection. Transp Res Part C Emerg Technol2020; 115: 18.
19.
CaiBRYangSQGaoL, et al. Hybrid variational autoencoder for time series forecasting. Knowl Based Syst2023; 281: 9.
20.
WangYXuXHuL, et al. Continuous imputation of missing values in time series via Wasserstein generative adversarial imputation networks and variational auto-encoders model. Physica A2024; 647: 129914.
21.
HarrouFDairiAZeroualA, et al. Forecasting of bicycle and pedestrian traffic using flexible and efficient hybrid deep learning approach. Appl Sci2022; 12(9): 4482.
22.
LiuJWHuangYCWuD, et al. Multi-Channel Multi-Scale Convolution Attention Variational Autoencoder (MCA-VAE): an interpretable anomaly detection algorithm based on variational autoencoder. Sensors2024; 24(16): 24.
23.
MushtaqAHaqIUSarwarMA, et al. Multi-agent reinforcement learning for traffic flow management of autonomous vehicles. Sensors2023; 23(5): 2373.
24.
AgarwalASahuDMohataR, et al. Dynamic traffic signal control for heterogeneous traffic conditions using max pressure and reinforcement learning. Expert Syst Appl2024; 254: 124416.
25.
ShaoZZhangZWangF, et al. Spatial-temporal identity: a simple yet effective baseline for multivariate time series forecasting. In: 31st ACM international conference on information and knowledge management, Atlanta, GA, USA, 17–21 October 2022, pp.4454–4458. New York: ACM.
26.
LiQHanZWuXM.Deeper insights into graph convolutional networks for semi-supervised learning. Proc AAAI Conf Artif Intell2018; 32(1): 3538–3545.
ZhuCFYuCXHuoJ.Research on spatio-temporal network prediction model of parallel-series traffic flow based on Transformer and GCAT. Physica A2023; 610: 20.
SongCLinYGuoS, et al. Spatial-temporal synchronous graph convolutional networks: a new framework for spatial-temporal network data forecasting. Proc AAAI Conf Artif Intell2020; 34(1): 914–921.
31.
ChoiJChoiHHwangJ, et al. Graph neural controlled differential equations for traffic forecasting. Proc AAAI Conf Artif Intell2022; 36: 6367–6374.
