A Short-Term Traffic Flow Prediction to Control Traffics in Large Scale Transportation using Internet of Things
S. Saravanan1, K. Venkatachalapathy2
1S. Saravanan, Assistant Professor/Programmer, Department of Computer and Information Science, Annamalai University, Chidambaram, India,
2K. Venkatachalapathy, Professor, Department of Computer and Information Science, Annamalai University, Chidambaram, India.
Manuscript received on November 12, 2019. | Revised Manuscript received on November 23, 2019. | Manuscript published on 30 November, 2019. | PP: 8323-8330 | Volume-8 Issue-4, November 2019. | Retrieval Number: D8988118419/2019©BEIESP | DOI: 10.35940/ijrte.D8988.118419
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Traffic congestion is the key problem that occurs across urban metropolises around the world. Due to the increase in transportation vehicles the fixed light time on traffic signals not able to solve the traffic congestion problem. In this paper, First, we develop an IoT based system which is capable of streaming the traffic surveillance footages to cloud storage, then the vehicle count is recorded every 30 sec interval and updated in the traffic flow dataset. Second the traffic flow is predicted using our CNN-LSTM residual learning model. Finally, the predicted value is classified and traffic density at each road section is identified, thereby passing this density value to green light time calculation to set an optimal green time to reduce the traffic congestion. The traffic flow dataset, China is used for training and testing to forecast the short time traffic flow across the road section. Experiment results shows that our model has best accuracy by lowering the RMSE value.
Keywords: CNN, LSTM, Prediction, Traffic flow.
Scope of the Article: Internet of Things.