Urban Traffic Flow Forecast Based on FastGCRNN

Traffic forecasting is an important prerequisite for the application of intelligent transportation systems in urban traffic networks. The existing works adopted RNN and CNN/GCN, among which GCRN is the state-of-the-art work, to characterize the temporal and spatial correlation of traffic flows. However, it is hard to apply GCRN to the large-scale road networks due to high computational complexity. To address this problem, we propose abstracting the road network into a geometric graph and building a Fast Graph Convolution Recurrent Neural Network (FastGCRNN) to model the spatial-temporal dependencies of traffic flow. Specifically, we use FastGCN unit to efficiently capture the topological relationship between the roads and the surrounding roads in the graph with reducing the computational complexity through importance sampling, combine GRU unit to capture the temporal dependency of traffic flow, and embed the spatiotemporal features into Seq2Seq based on the Encoder-Decoder framework. Experiments on large-scale traffic data sets illustrate that the proposed method can greatly reduce computational complexity and memory consumption while maintaining relatively high accuracy.

[1]  Yan Tian,et al.  Traffic flow prediction using LSTM with feature enhancement , 2019, Neurocomputing.

[2]  Ken-ichi Kawarabayashi,et al.  Representation Learning on Graphs with Jumping Knowledge Networks , 2018, ICML.

[3]  Maninder Singh Setia,et al.  Methodology Series Module 5: Sampling Strategies , 2016, Indian journal of dermatology.

[4]  Naixue Xiong,et al.  Context-aware cross-layer optimized video streaming in wireless multimedia sensor networks , 2010, The Journal of Supercomputing.

[5]  T. Chai,et al.  Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature , 2014 .

[6]  Naixue Xiong,et al.  Design and Analysis of Probing Route to Defense Sink-Hole Attacks for Internet of Things Security , 2020, IEEE Transactions on Network Science and Engineering.

[7]  Alexander Mendiburu,et al.  A Review of Travel Time Estimation and Forecasting for Advanced Traveler Information Systems , 2012 .

[8]  Naixue Xiong,et al.  A Decentralized and Adaptive Flocking Algorithm for Autonomous Mobile Robots , 2008, 2008 The 3rd International Conference on Grid and Pervasive Computing - Workshops.

[9]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[10]  Naixue Xiong,et al.  A novel dynamic network data replication scheme based on historical access record and proactive deletion , 2012, The Journal of Supercomputing.

[11]  A. R. Cook,et al.  ANALYSIS OF FREEWAY TRAFFIC TIME-SERIES DATA BY USING BOX-JENKINS TECHNIQUES , 1979 .

[12]  Jie Wu,et al.  A Self-tuning Failure Detection Scheme for Cloud Computing Service , 2012, 2012 IEEE 26th International Parallel and Distributed Processing Symposium.

[13]  Junqiang Leng,et al.  An adaptive hybrid model for short-term urban traffic flow prediction , 2019, Physica A: Statistical Mechanics and its Applications.

[14]  Naixue Xiong,et al.  On Studying the Impact of Uncertainty on Behavior Diffusion in Social Networks , 2015, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[15]  Naixue Xiong,et al.  Adaptive unequal protection for wireless video transmission over IEEE 802.11e networks , 2013, Multimedia Tools and Applications.

[16]  Gary A. Davis,et al.  Nonparametric Regression and Short‐Term Freeway Traffic Forecasting , 1991 .

[17]  Naixue Xiong,et al.  Intelligent model design of cluster supply chain with horizontal cooperation , 2012, J. Intell. Manuf..

[18]  Naveen Kumar Chikkakrishna,et al.  Short-Term Traffic Prediction Using Sarima and FbPROPHET , 2019, 2019 IEEE 16th India Council International Conference (INDICON).

[19]  Cyrus Shahabi,et al.  Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting , 2017, ICLR.

[20]  Naixue Xiong,et al.  An Energy-Efficient Dynamic Power Management in Wireless Sensor Networks , 2006, 2006 Fifth International Symposium on Parallel and Distributed Computing.

[21]  Samy Bengio,et al.  Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks , 2019, KDD.

[22]  Naixue Xiong,et al.  Using Multi-Modal Semantic Association Rules to fuse keywords and visual features automatically for Web image retrieval , 2011, Inf. Fusion.

[23]  Xiangyan Tang,et al.  Traffic Flow Combination Forecasting Method Based on Improved LSTM and ARIMA , 2019, Int. J. Embed. Syst..

[24]  Naixue Xiong,et al.  Efficient Protocols for Privacy Preserving Matching Against Distributed Datasets , 2006, ICICS.

[25]  Junbo Zhang,et al.  Urban Traffic Prediction from Spatio-Temporal Data Using Deep Meta Learning , 2019, KDD.

[26]  Cao Xiao,et al.  FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling , 2018, ICLR.

[27]  Xianfeng Song,et al.  Matching of vehicle GPS traces with urban road networks , 2010 .

[28]  I Okutani,et al.  Dynamic prediction of traffic volume through Kalman Filtering , 1984 .

[29]  Aura Reggiani,et al.  Introduction: Cross Atlantic Perspectives in Methods and Models Analysing Transport and Telecommunications , 2005 .

[30]  Naixue Xiong,et al.  A sustainable heuristic QoS routing algorithm for pervasive multi-layered satellite wireless networks , 2010, Wirel. Networks.

[31]  Naixue Xiong,et al.  A Kernel-Based Compressive Sensing Approach for Mobile Data Gathering in Wireless Sensor Network Systems , 2018, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[32]  Joan Bruna,et al.  Spectral Networks and Locally Connected Networks on Graphs , 2013, ICLR.

[33]  Tao Lin,et al.  Estimating Traffic Flow in Large Road Networks Based on Multi-Source Traffic Data , 2021, IEEE Transactions on Intelligent Transportation Systems.

[34]  Naixue Xiong,et al.  Connectivity and coverage maintenance in wireless sensor networks , 2010, The Journal of Supercomputing.

[35]  Antoine G. Hobeika,et al.  A short-term demand forecasting model from real-time traffic data , 1993 .

[36]  Xavier Bresson,et al.  Structured Sequence Modeling with Graph Convolutional Recurrent Networks , 2016, ICONIP.

[37]  Quoc V. Le,et al.  Sequence to Sequence Learning with Neural Networks , 2014, NIPS.

[38]  Tao Cheng,et al.  A graph deep learning method for short‐term traffic forecasting on large road networks , 2019, Comput. Aided Civ. Infrastructure Eng..

[39]  Changxi Ma,et al.  Short-Term Traffic Flow Prediction Method for Urban Road Sections Based on Space–Time Analysis and GRU , 2019, IEEE Access.

[40]  Naixue Xiong,et al.  Dynamic power management in new architecture of wireless sensor networks , 2009, Int. J. Commun. Syst..

[41]  Xianglong Luo,et al.  An Algorithm for Traffic Flow Prediction Based on Improved SARIMA and GA , 2018, KSCE Journal of Civil Engineering.

[42]  Naixue Xiong,et al.  Distributed k-connected fault-tolerant topology control algorithms with PSO in future autonomic sensor systems , 2012, Int. J. Sens. Networks.

[43]  Xiao-Ming Wu,et al.  Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning , 2018, AAAI.

[44]  Pengcheng Li,et al.  Prediction of Taxi Demand Based on ConvLSTM Neural Network , 2018, ICONIP.