Journals Books 2687-5527 doi.org/10.36287/setsci
Latest Issue Archive Future Issues About Us
Conference Proceedings

SETSCI - Volume 4 (6) (2019)
ISAS WINTER-2019 (ENS) - 4th International Symposium on Innovative Approaches in Engineering and Natural Sciences, Samsun, Turkey, Nov 22, 2019

Analysis of Highway Traffic Using Deep Learning Techniques / Derin Öğrenme Teknikleri Kullanılarak Anayol Trafik Analizi
Muhammet Esad Özdağ1*, Nesrin Aydın Atasoy2
1Tokat Gaziosmanpaşa University, Tokat, Turkey
2Karabük University, Karabük, Turkey
* Corresponding author: muhammetesat.ozdag@gop.edu.tr
Published Date: 2019-12-22   |   Page (s): 384-390   |    221     10
https://doi.org/10.36287/setsci.4.6.098

ABSTRACT Traffic flow forecasting has an important place in designing a successful intelligent transportation system. The success of the forecasting is related to the accuracy and timely acquisition of the traffic flow data. The inadequacy in the number of data has led to the use of shallow architectures in the traffic forecasting models realized so far or to design models with generated data. These models failed to produce forecast results with sufficient success. Nowadays, in the age of big data, in parallel with the increase in traffic density, there has been a significant increase in the diversity and size of the collected traffic flow data. This result constitutes the main motivation in our study. Our study aims to forecast traffic density at the exit of a motorway that have linked roads. The forecasting models proposed in our study were designed using generally accepted, Deep Learning techniques, which can produce meaningful prediction results with big data. The techniques used in our study are Long-Short Time Memory (LSTM), Vanilla LSTM, Stacked LSTM, Bidirectional LSTM and Gated Recurrent Unit (GRU) neural networks. The dataset used in the study consists of 929 thousand 640 measurement data collected by loop sensors placed at 6 different points. The dataset was divided into two parts: 80% training and 20% testing. Forecast achievements of the designed models on the test dataset were recorded by calculating the Mean Square Error (MSE) values. The results show that Deep Learning techniques in traffic flow forecasting with low MSE values produce successful results and can be used in traffic flow prediction models. (tr) Başarılı bir akıllı ulaşım sisteminin tasarlanmasında, trafik akış tahmini önemli bir yer edinmektedir. Tahminin başarısı, akış verisinin doğruluğu ve zamanında elde edilmesi ile ilişkilidir. Veri sayısındaki yetersizlik, şimdiye kadar gerçekleşen trafik tahmin modellerinde sığ mimarilerin kullanılmasına yada üretilmiş yapay ölçüm verileri ile modeller tasarlanmasına sebep olmuştur. Bu modeller yeterli başarıya sahip tahmin sonuçları üretememiştir. Büyük veri çağına girdiğimiz günümüzde, trafik yoğunluğundaki artışa paralel olarak, toplanan trafik verilerin çeşitliliği ve büyüklüğündede gözle görülür bir artış gerçekleşmiştir. Bu sonuç, çalışmamızdaki temel motivasyonu oluşturmaktadır. Çalışmamız, bağlantı yollarına sahip bir otoyolun çıkışındaki trafik yoğunluğunu önceden tahmin etmeyi hedeflemektedir. Çalışmamızda önerilen tahmin modelleri, büyük veriler ile eğitilerek anlamlı tahmin sonuçları üretebileceği genel kabul gören Derin Öğrenme teknikleri kullanılarak tasarlanmıştır. Çalışmamızda kullanılan bu teknikler sırası ile, Long-Short Time Memory (LSTM), Vanilla LSTM, Stacked LSTM, Bidirectional LSTM ve Gated Recurrent Unit (GRU) sinir ağlarıdır. Çalışmada kullanılan veri seti, 6 farklı noktaya yerleştirilmiş döngü sensörleri ile toplanmış 929 bin 640 ölçüm verisini içermektedir. Bu veri seti kamuya açık bir web sitesinden elde edilmiştir. Veri seti, %80'i eğitim, %20'i test olmak üzere 2 bölüme ayrılmıştır. Geliştirilen modellerin test veri seti üzerindeki tahmin başarıları Ortalama Karesel Hata (MSE) değerleri hesaplanarak kaydedilmiştir. Sonuçlar, düşük MSE değerleri ile trafik akış tahmininde derin öğrenme tekniklerinin başarılı sonuçlar ürettiği ve trafik akış tahmin modellerinde kullanılabileceğini göstermektedir.
KEYWORDS Deep Learning, RNN, Intelligent Transportation System, Traffic Flow Forecasting, Big Data
REFERENCES [1] M. vanderVoort, M. Dougherty, and S. Watson, “Combining Kohonen maps with ARIMA time series models to forecast traffic flow,” Transp. Res. C, Emerging Technol., vol. 4, no. 5, pp. 307–318, Oct. 1996.
[2] B. M. Williams, “Multivariate vehicular traffic flow prediction—Evaluation of ARIMAX modeling,” Transp. Res. Rec., no. 1776, pp. 194–200, 2001.
[3] S. Sun, C. Zhang, and Y. Guoqiang, “A Bayesian network approach to traffic flow forecasting,” IEEE Intell. Transp. Syst. Mag., vol. 7, no. 1, pp. 124–132, Mar. 2006.
[4] A. Lakhina, K. Papagiannaki, M. Crovella, C. Diot, E. D. Kolaczyk, and N. Taft, “Structural analysis of network traffic flows,” SIGMETRICS Perform. Eval. Rev., vol. 32, no. 1, pp. 61–72, 2004.
[5] Y. Kamarianakis and P. Prastacos, “Forecasting traffic flow conditions in an urban network—Comparison of multivariate and univariate approaches,” Transp. Res. Rec., no. 1857, pp. 74–84, Transporation Network Modeling 2003: Planning and Administration, 2003.
[6] E. I. Vlahogianni, M. G. Karlaftis, and J. C. Golias, “Optimized and metaoptimized neural networks for short-term traffic flow prediction: A genetic approach,” Transp. Res. C, Emerging Technol., vol. 13, no. 3, pp. 211–234, Jun. 2005.
[7] K. Y. Chan, T. S. Dillon, J. Singh, and E. Chang, “Neural-network-based models for short-term traffic flow forecasting using a hybrid exponential smoothing and Levenberg–Marquardt algorithm,” IEEE Trans. Intell. Transp. Syst., vol. 13, no. 2, pp. 644–654, Jun. 2012.
[8] X. Jiang and H. Adeli, “Dynamic wavelet neural network model for traffic flow forecasting,” J. Transp. Eng., vol. 131, no. 10, pp. 771–779, Oct. 2005.
[9] N. G. Polson and V. O. Sokolov, “Deep learning for short-term traffic flow prediction,” Transp. Res. C, Emerg. Technol., vol. 79, pp. 1–17, Jun. 2017
[10] W. Z. Zheng, D. H. Lee, and Q. X. Shi, “Short-term freeway traffic flow prediction: Bayesian combined neural network approach,” J. Transp. Eng., vol. 132, no. 2, pp. 114–121, Feb. 2006.
[11] B. M. Williams, P. K. Durvasula, and D. E. Brown, “Urban freeway traffic flow prediction—Application of seasonal autoregressive integrated moving average and exponential smoothing models,” Transp. Res. Rec., no. 1644, pp. 132–141, 1998.
[12] M. Castro-Neto, Y.-S. Jeong, M.-K. Jeong, and L. D. Han, “Online-SVR for short-term traffic flow prediction under typical and atypical traffic conditions,” Expert Syst. Appl., vol. 36, no. 3, pp. 6164–6173, Apr. 2009.
[13] T. Maze, M. Agarwai, and G. Burchett, “Whether weather matters to traffic demand, traffic safety, and traffic operations and flow,” Transportation research record: Journal of the transportation research board, no. 1948, pp. 170–176, 2006.
[14] M. Lippi, M. Bertini, and P. Frasconi, “Short-term traffic flow forecasting: An experimental comparison of time-series analysis and supervised learning,” IEEE Trans. Intell. Transp. Syst., vol. 14, no. 2, pp. 871–882, Jun. 2013.
[15] Lv, Y., Duan, Y., Kang, W., et al.: ‘Traffic flow prediction with big data: A deep learning approach’, IEEE Trans. Intell. Transp. Syst., 16, (2), pp. 865–873, 2015.
[16] F. Chollet and J. J. Allaire, Deep Learning with R, Shelter Island, NY, USA: Manning Publications Company, 2018.
[17] N. Srivastava et al., “Dropout: A simple way to prevent neural networks from overfitting,” J. Mach. Learn. Res., vol. 15, no. 1, pp. 1929–1958, 2014.
[18] N. Ketkar, Deep Learning with Python a Hands-on Introduction, 1st ed. Apress, 2017.
[19] J. Patterson and A. Gibson, Deep Learning: A Practitioner’s Approach, O’Reilly Media, Inc., 2017.
[20] N. Buduma and N. Locascio, Fundamentals of Deep Learning: Designing Next-generation Machine Intelligence Algorithms, O’Reilly Media, Inc., 2017.
[21] S. Heykin, Neural Networks and Learning Machines (3rd Editiom), Pearson, New Jersey, 2009.
[22] S. Pattanayak, Pro Deep Learning with Tensorflow: A Mathematical Aproach to Advanced Artificial Intelligence in Python, Apress Media, pp.153-278, 2017.
[23] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.
[24] P. Goyal, S. Pandey, and K. Jain, Deep Learning for Natural Language Processing, Apress, 2018.
[25] J. Chung, Ç. Gülçehre, K. Cho and Y. Bengio, “Empirical evaluation of gated recurrent neural networks on sequence modeling,” CoRR, abs/1412.3555, 2014.
[26] H. Demuth, M. Beale and M. Hagan, User Guide, Neural Networks ToolboxTM 6, The MathWorks, Inc., ISBN:0-9717321-0-8, 2-25, 3-5, 8-6, 2009


SET Technology - Turkey

eISSN  : 2687-5527    
DOI : doi.org/10.36287/setsci

E-mail : info@set-science.com
+90 533 2245325

Tokat Technology Development Zone Gaziosmanpaşa University Taşlıçiftlik Campus, 60240 TOKAT-TURKEY
©2018 SET Technology