Vehicle Traffic Flow Forecasting on Caltrans PeMS Dataset Using Machine Learning Algorithms and LSTM Networks
Authors: 1 Jiarui Chang, 2 Jingwen Du, 3 Hojin Chung
Affiliation: 1 Rice University (Houston, USA), 2 Cornell University (Ithaca, USA) , 3 Gyeonggi Suwon International School (Suwon, Korea)
Category:
Keywords: Deep learning model, Traffic flow prediction, Caltrans PeMs dataset, LSTM model
ABSTRACT. In Intelligent transportation systems, accurate traffic flow prediction is fundamental in transportation modeling and management.Previous studies have classified prediction approaches into three categories including a time series approach with ARIMA model for finding traffic flow patterns and using those patterns for prediction, a probabilistic approach for modeling and forecasting from a probabilistic perspective, and nonparametric approaches that can perform better by handling un-deterministic and complex time series traffic datasets.This paper analyzes historical timeseries traffic data from sensors using machine learning algorithms as baseline models and designs a deep learning LSTM model to train using the historical dataset to forecast traffic flow using the trained model. The paper also compares the performance of machine learning algorithms and the deep learning model. The results show the deep learning LSTM model to outperform machine learning models.
References:
Dai, X., Fu, R., Lin, Y., Li, L., & Wang, F. Y. (2017). DeepTrend: A deep hierarchical neural network for traffic flow prediction. arXiv preprint arXiv:1707.03213
Chen, W., An, J., Li, R., Fu, L., Xie, G., Bhuiyan, M. Z. A., & Li, K. (2018). A novel fuzzy deep-learning approach to traffic flow prediction with uncertain spatial–temporal data features. Future Generation Computer Systems, 89, 78-88
Manornjitham, Raj. P, Lal, H.K (2018). A Survey of Road Traffic Prediction with deep learning. International Journal of Pure and Applied Mathematics, 2065-2073
Jia, Y., Wu, J., & Xu, M. (2017). Traffic flow prediction with rainfall impact using a deep learning method. Journal of advanced transportation, 2017
Yang, B., Sun, S., Li, J., Lin, X., & Tian, Y. (2019). Traffic flow prediction using LSTM with feature enhancement. Neurocomputing, 332, 320-327
S.K.Groen. (2012), A Review of Traffic growth rate calculations. Shaping the future: Linking policy, research and outcomes, 25th ARRB Conference, Perth, Australia, 1-15
Polson, N. G., & Sokolov, V. O. (2017). Deep learning for short-term traffic flow prediction. Transportation Research Part C: Emerging Technologies, 79, 1-17
Du, S., Li, T., Gong, X., Yu, Z., Huang, Y., &Horng, S. J. (2018). A hybrid method for traffic flow forecasting using multimodal deep learning. arXiv preprint arXiv:1803.02099
Lv, Z., Xu, J., Zheng, K., Yin, H., Zhao, P., & Zhou, X. (2018, January). LC-RNN: A Deep Learning Model for Traffic Speed Prediction. In IJCAI (pp. 3470-3476)
Wang, J., Chen, R., & He, Z. (2019). Traffic speed prediction for urban transportation network: A path based deep learning approach. Transportation Research Part C: Emerging Technologies, 100, 372-385
Zhang, S., Kang, Z., Hong, Z., Zhang, Z., Wang, C., & Li, J. (2018, July). Traffic flow prediction based on cascaded artificial neural network. In IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium(pp. 7232-7235). IEEE
Xiao, Y., & Yin, Y. (2019). Hybrid LSTM Neural Network for Short-Term Traffic Flow Prediction. Information, 10(3), 105
Tian, Y., Zhang, K., Li, J., Lin, X., & Yang, B. (2018). LSTM-based traffic flow prediction with missing data. Neurocomputing, 318, 297-305
Menu