Short‐Term Traffic Speed Forecasting Based on Attention Convolutional Neural Network for Arterials

Liu, Qingchao; Wang, Bo-Chen; Zhu, Yuquan

doi:10.1111/mice.12417

Cited by 138 publications

(78 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Attention mechanism has been widely utilized in traffic prediction tasks to improve the performance of the models, [37], [38]. With attention block, one model can focus on important features and suppressing unnecessary ones [39].…”

Section: Convolutional Block Attention Modulementioning

confidence: 99%

Short-Term Traffic Speed Prediction of Urban Road With Multi-Source Data

2020

View full text Add to dashboard Cite

In recent years, the convolutional and recurrent neural networks are widely applied in traffic prediction tasks. Traffic speed prediction is an important and challenging topic in intelligent transportation systems. In this case, this paper proposes a hybrid deep learning structure for short-term traffic speed prediction, which combines convolutional neural networks and long short-term memory neural networks together. External factors such as weather condition and air quality can also affect the driving behavior of travelers and cause fluctuation of traffic speed. Based on theories in traffic engineering, we propose a data-fusion method to measure the impact of environmental factors. To enhance the performance of our model, we introduced attention mechanism to our model. With convolutional block attention module, our network could emphasize important channels and pixels of input features and suppress unnecessary ones. Comparing with several deep learning methods and hybrid deep learning structures, an experiment in one region of Suzhou which contains 909 links shows the outperformance of our model. Under different time steps, the prediction error of our model is lower than any other methods in urban expressway, primaryarterial, secondary-arterial, and branch-road. The results indicate that the spatial dependencies, the temporal correlations, and environmental impact should not be ignored in traffic speed prediction tasks.

show abstract

Section: Convolutional Block Attention Modulementioning

confidence: 99%

Short-Term Traffic Speed Prediction of Urban Road With Multi-Source Data

2020

View full text Add to dashboard Cite

show abstract

“…The results showed that the proposed method outperformed LSTM model by a mean squared errors improvement of 42.91%. Using the traffic speed data from the Caltrans Performance Measurement System (PeMS), Liu et al [43] predicted traffic speed by the attention convolutional neural network (ACNN) model and found that the proposed model achieved better forecast results than traditional linear models.…”

Section: Short-term Traffic Speed Predictionmentioning

confidence: 99%

Short-term traffic speed prediction under different data collection time intervals using a SARIMA-SDGM hybrid prediction model

Song

et al. 2019

PLoS ONE

View full text Add to dashboard Cite

Short-term traffic speed prediction is a key component of proactive traffic control in the intelligent transportation systems. The objective of this study is to investigate the short-term traffic speed prediction under different data collection time intervals. Traffic speed data was collected from an urban freeway in Edmonton, Canada. A seasonal autoregressive integrated moving average plus seasonal discrete grey model structure (SARIMA-SDGM) was proposed to perform the traffic speed prediction. The model performance of SARIMA-SDGM model was compared with that of the seasonal autoregressive integrated moving average (SARIMA) model, seasonal discrete grey model (SDGM), artificial neural network (ANN) model, and support vector regression (SVR) model. The results showed that SAR-IMA-SDGM model performs best with the lowest mean absolute error (MAE), mean absolute percentage error (MAPE), and the root mean square error (RMSE). The traffic speed prediction accuracy under different time intervals were compared based on the SARIMA-SDGM model. The results showed that the prediction accuracy improves with the increase in time interval. In addition, when the time interval is greater than 10 min, the prediction results yield stable prediction accuracy.

show abstract

“…The attention allows for more direct dependence between the states of the model at different time steps by integrating with other deep models, eg, CNNs or RNNs. Liu et al proposed an attention‐based CNN model for traffic speed prediction, which employed CNN to extract spatial and temporal features of traffic data, and simultaneously utilized attention to weight feature maps and channels to enhance the validity of the learned features. Moreover, the attention‐based RNNs have been successfully applied to sequence learning and traffic prediction .…”

Section: Related Workmentioning

confidence: 99%

Spatiotemporal traffic matrix prediction: A deep learning approach with wavelet multiscale analysis

Zhao

et al. 2019

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Network traffic analysis has always been a key technique for operating and managing a network. However, due to some (non) technical issues, it is not trivial to directly obtain network-wide traffic data. Although a large number of traffic matrix (TM) prediction methods have been used to obtain future network-wide traffic, they achieve somewhat limited accuracy due to neglecting spatiotemporal evolution features of TM series at different time scales. In order to improve the performance of TM prediction, we propose a novel end-to-end deep neural network based on wavelet multiscale analysis, called WSTNet. In this network, the original TM series is first decomposed into multilevel time-frequency TM subseries at different time scales by using discrete wavelet decomposition, and then the convolutional neural network without pooling is used to extract the spatial patterns among traffic flows, and finally, the long short-term memory neural network with a self-attention mechanism by relating different positions of input sequences across entire time steps is employed to explore the temporal evolution features within TM series. To investigate the performance of our proposed model, extensive experiments are conducted on two real network traffic data sets from the Abilene and GÉANT backbone networks. The results show that WSTNet is significantly better than the other four state-of-the-art deep learning methods.Trans Emerging Tel Tech. 2019;30:e3640.wileyonlinelibrary.com/journal/ett

show abstract

Short‐Term Traffic Speed Forecasting Based on Attention Convolutional Neural Network for Arterials

Cited by 138 publications

References 41 publications

Short-Term Traffic Speed Prediction of Urban Road With Multi-Source Data

Short-Term Traffic Speed Prediction of Urban Road With Multi-Source Data

Short-term traffic speed prediction under different data collection time intervals using a SARIMA-SDGM hybrid prediction model

Spatiotemporal traffic matrix prediction: A deep learning approach with wavelet multiscale analysis

Contact Info

Product

Resources

About