Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks

By Haiyang Yu, Zhihai Wu, Shuqin Wang, Yunpeng Wang, Xiaolei Ma. Italian National Conference on Sensors, 2017

594 citations recorded by the source corpus.

Predicting large-scale transportation network traffic has become an important and challenging topic in recent decades. Inspired by the domain knowledge of motion prediction, in which the future motion of an object can be predicted based on previous scenes, we propose a network grid representation method that can retain the fine-scale structure of a transportation network. Network-wide traffic speeds are converted into a series of static images and input into a novel deep architecture, namely, spatiotemporal recurrent convolutional networks (SRCNs), for traffic forecasting. The proposed SRCNs inherit the advantages of deep convolutional neural networks (DCNNs) and long short-term memory (LSTM) neural networks. The spatial dependencies of network-wide traffic can be captured by DCNNs, and the temporal dynamics can be learned by LSTMs. An experiment on a Beijing transportation network with 278 links demonstrates that SRCNs outperform other deep learning-based algorithms in both short-term and long-term traffic prediction.

Subfields: Law, Political Science and International Relations, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/bbd7a9cfb0b28fe550c9bfd59d0d1c9d05ac1588

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Dynamic pricing and matching in ride‐hailing platforms (2019, unknown): “A variety of forecasting techniques have been applied to these problems, including time-series analysis and machine learning (Yu et al. 2017, Yao et al. 2018).”
A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting (2020, methodology): “Yu et al. [36] captured spatial correlation and temporal dynamics by using DCNN and LSTM, respectively.”
High-Order Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting (2018, background): “l structure in a physical traffic network. Moreover, interpreting the network parameters in terms of real-world spatial dependencies is most often impossible. To address this, other works [13], [14], [15] attempt to model spatial dependencies with convolutional neural network (CNN). However, convent”
T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction (2018, methodology): “[44] used the Deep Convolutional Neural Network(DCNN) to capture spatial dependence, used LSTM to capture temporal dynamics, and demonstrated the superiority of the SRCN model through experiments on Beijing traffic network data.”
Graph Neural Networks for Intelligent Transportation Systems: A Survey (2023, methodology): “Finally, they validated the effectiveness of their methodology using three ridership datasets over the state-of-the-art baseline models: LSTM, CNN, GCN, DMVST-NET [55], CNN-LSTM [80], SRCNs [87], SBULSTM [88], DCRNN [6], STGCN [27], Graph-WaveNet [34], and PVCGN [85].”
Stacked Bidirectional and Unidirectional LSTM Recurrent Neural Network for Forecasting Network-wide Traffic State with Missing Values (2020, methodology): “Although some studies Chen et al. (2016); Wu and Tan (2016); Yu et al. (2017b) utilized more than one LSTM layers, the inﬂuence of the number of LSTM layers needs to be further evaluated.”
Optimized Graph Convolution Recurrent Neural Network for Traffic Prediction (2021, methodology): “Thus the spatial relations were modeled by matrices or nonlinear functions [26], for example, CNN [37], [39].”
A Survey on Modern Deep Neural Network for Traffic Prediction: Trends, Methods and Challenges (2020, methodology): “data is used in:, [4], [54], [62], [65] and [67].”