Deep Bidirectional and Unidirectional LSTM Recurrent Neural Network for Network-wide Traffic Speed Prediction

By Zhiyong Cui, Ruimin Ke, Yinhai Wang. 2018

464 citations recorded by the source corpus.

Short-term traffic forecasting based on deep learning methods, especially long short-term memory (LSTM) neural networks, has received much attention in recent years. However, the potential of deep learning methods in traffic forecasting has not yet fully been exploited in terms of the depth of the model architecture, the spatial scale of the prediction area, and the predictive power of spatial-temporal data. In this paper, a deep stacked bidirectional and unidirectional LSTM (SBU- LSTM) neural network architecture is proposed, which considers both forward and backward dependencies in time series data, to predict network-wide traffic speed. A bidirectional LSTM (BDLSM) layer is exploited to capture spatial features and bidirectional temporal dependencies from historical data. To the best of our knowledge, this is the first time that BDLSTMs have been applied as building blocks for a deep architecture model to measure the backward dependency of traffic data for prediction. The proposed model can handle missing values in input data by using a masking mechanism. Further, this scalable model can predict traffic speed for both freeway and complex urban traffic networks. Comparisons with other classical and state-of-the-art models indicate that the proposed SBU-LSTM neural network achieves superior prediction performance for the whole traffic network in both accuracy and robustness.

Subfields: Building and Construction, Control and Systems Engineering, Transportation

View Paper: https://www.semanticscholar.org/paper/696b3fa5e6b602b5e9f41a08424f543f87906431

How researchers see this paper

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

In Building and Construction

In Control and Systems Engineering

In Transportation

Citations in the corpus

A survey on long short-term memory networks for time series prediction (2021, background): “This is tested and evaluated by [11] where bidirectional LSTM networks encapsulate spatially and temporally distributed information and can handle incomplete data by a flexible connection mechanism for the propagation of the cell state vector.”
Graph Neural Networks for Intelligent Transportation Systems: A Survey (2023, methodology): “Unlike many previous studies, the output of the GCN and SBULSTM modules are parallelly calculated and concatenated to avoid the distortion of temporal patterns (which occurs in ordinary sequential CNN-LSTM frameworks).”
Dynamic Spatial-Temporal Graph Convolutional Neural Networks for Traffic Forecasting (2019, methodology): “We first measure the impact of spatial order of the input data on the learning of spatial dependencies, as done in (Cui, Ke, and Wang 2018).”
Stacked Bidirectional and Unidirectional LSTM Recurrent Neural Network for Forecasting Network-wide Traffic State with Missing Values (2020, background): “Existing studies Graves et al. (2013); LeCun et al. (2015) have shown that LSTM architectures with several hidden layers can progressively build up a higher level of representations of sequence data, and thus, work more eﬀectively.”
High-Order Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting (2018, background): “RNN and its variants (long short-term memory (LSTM) recurrent neural network and gated recurrent unit (GRU)) show excellent potential for traffic forecasting [4, 15, 24].”
A Deep Blockchain Framework-Enabled Collaborative Intrusion Detection for Protecting IoT and Cloud Networks (2021, methodology): “…hidden layer −→ h ( t ) and the output layer y ( t ) by reiterating the forward layer starting t = 1 to t f , backward hidden layer since t = t f to 1, and then updating the ﬁnal value using the following equations [29]: Authorized licensed use limited to: University of Texas at San Antonio.”
Bi-Directional ConvLSTM U-Net with Densley Connected Convolutions (2019, background): “It has been proved that analyzing both forward and backward temporal perspectives enhanced the predictive performance [9].”
A Survey of Traffic Prediction: from Spatio-Temporal Data to Intelligent Transportation (2021, background): “[168] propose a deep stacked bidirectional and unidirectional LSTM neural network architecture, which considers both forward and backward dependencies in time series data, to predict network-wide traffic speed.”
STREETS: A Novel Camera Network Dataset for Traffic Flow (2019, background): “The work in [23, 24, 27] all use the aforementioned PeMS data source [22], while [28] and [29] use in-roadway sensors around Beijing and Seattle, respectively.”
A Survey on Modern Deep Neural Network for Traffic Prediction: Trends, Methods and Challenges (2020, methodology): “Additionally, these models were proposed at a time where traffic data were simpler and much smaller in size [23], a condition that no longer”
Process (2020, background): “Most of the previous works [10, 11, 12, 13, 14, 15, 16, 17, 18, 19] on traffic modeling focus on predicting speed, volume, density and travel time, which have achieved remarkable success in this field.”
Traffic Graph Convolutional Recurrent Neural Network: A Deep Learning Framework for Network-Scale Traffic Learning and Forecasting (2018, background): “Due to their efficient use of temporal dependencies, RNN and its variants (long short-term memory (LSTM) [9] and gated recurrent unit (GRU) [10] recurrent neural networks) show excellent potential for traffic forecasting [5], [11], [12].”