Speaker
Description
Abstract
This study opens new directions on Traffic Modeling using Neural Networks, their limitations and exploring existing research on the use of FNN, RNN and CNN architectures in this domain. Our proposed hybrid model leverages the strengths of both CNNs for recognizing spatial patterns and RNNs for capturing temporal dependencies in traffic data. The paper details our methodology, including data collection, network architecture design, training process, hyperparameter tuning, and performance evaluation. We compare our results with traditional methods and discuss their implications for intelligent transportation systems (ITS) and urban planning.
Keywords: Safety Distance, Braking Performance, Microcontroller, Sensors.
Reference:
[1] Ferriol-Galmés, M., Paillisse, J., Suárez-Varela, J., Rusek, K., Xiao, S., Shi, X., ... & Cabellos-Aparicio, A. (2023). RouteNet-Fermi: Network modeling with graph neural networks. IEEE/ACM transactions on networking, 31(6), 3080-3095.
[2] Chan, R. K. C., Lim, J. M. Y., & Parthiban, R. (2021). A neural network approach for traffic prediction and routing with missing data imputation for intelligent transportation system. Expert Systems With Applications, 171, 114573.
[3] Abdullah, S. M., Periyasamy, M., Kamaludeen, N. A., Towfek, S. K., Marappan, R., Kidambi Raju, S., Alharbi, A. H., & Khafaga, D. S. (2023). Optimizing Traffic Flow in Smart Cities: Soft GRU-Based Recurrent Neural Networks for Enhanced Congestion Prediction Using Deep Learning. Sustainability, 15(7), 5949.
