Prediction of traffic risks based on a neural network
DOI:
https://doi.org/10.26906/SUNZ.2025.1.10-16Keywords:
traffic safety, risk factors, modeling, vehicle, neural network, intelligent transport systemAbstract
On the basis of the structural analysis of traffic accident statistics in Ukraine, the most dangerous causes were identified: exceeding the safe speed, violation of maneuvering rules, violation of the rules for crossing intersections and pedestrian crossings. The proposed models for predicting the danger of road traffic and the risk of driving are considered. The factors of road traffic danger have been studied and systematized. A traffic risk prediction model is proposed using an intelligent transport system (ITS) and a neural network. The model is built on the technology of neural network processing of weighted statistical and dynamic arrays of input data characterizing the internal and external environment of the vehicle in order to obtain a traffic risk assessment. A neural network risk prediction algorithm is proposed. The learning of a multilayer neural network is carried out using the backpropagation algorithm. Testing of the forecasting system demonstrated the accuracy of predictions of 85-90%.
Downloads
References
1. Офіційний сайт патрульної поліції України. Статистика. URL: https://patrolpolice.gov.ua/statystyka/
2. Положий Д. С., Орєхов О. О. Інтелектуальні системи автомобільної безпеки на основі хмарних архітектур // Системи управління, навігації та зв'язку, 2023. № 4. С. 91-95. DOI: https://doi.org/10.26906/SUNZ.2023.4.091
3. Кашканов А.А., Пальчевський О.В. Інтелектуалізація управління дорожнім рухом як засіб підвищення ефективності транспортної мережі міста в неординарних ситуаціях // DOI: https://doi.org/10.31649/2413-4503-2022-16-2-42-50 DOI: https://doi.org/10.31649/2413-4503-2022-16-2-42-50
4. Anukruthi Karre, Akramul Azim. Collision Prediction of Smart Cities Using Co-simulation // Procedia Computer Sciance 224 (2023) 205-214. URL: www.sciencedirect.com DOI: https://doi.org/10.1016/j.procs.2023.09.029
5. Hou J., List G. F., Guo X. 2014. New Algorithms for Computing the Time-to-Collision in Freeway Traffic Simulation Models URL: https://doi.org/10.1155/2014/761047 DOI: https://doi.org/10.1155/2014/761047
6. Sharma, B.; Katiyar, V.K.; Kumar, K. Kranti Kumar Traffic Accident Prediction Model Using Support Vector Machines with Gaussian Kernel. In Proceedings of Fifth International Conference on Soft Computing for Problem Solving; Pant, M., Deep, K., Bansal, J.C., Nagar, A., Das, K.N., Eds.; Springer: Singapore, 2016; pp. 1–10. DOI: https://doi.org/10.1007/978-981-10-0451-3_1
7. Wu, Q.; Hui, L.C.K.; Yeung, C.Y.; Chim, T.W. Early car collision prediction in VANET. In International Conference on Connected Vehicles and Expo (ICCVE); IEEE: Shenzhen, China, 2015; pp. 19–23. DOI: https://doi.org/10.1109/ICCVE.2015.55
8. Aung N., Zhang, W., Dhelim, S., Yibo Ai. Accident Prediction System Based on Hidden Markov Model for Vehicular AdHoc Network in Urban Environments //MDPI, 2018, 9, 311. doi:10.3390/info9120311 DOI: https://doi.org/10.3390/info9120311
9. Strickland, M., Fainekos, G., & Ben Amor, H. (2018). Deep predictive models for collision risk assessment in autonomous driving. In 2018 IEEE International Conference on Robotics and Automation, ICRA 2018 (pp. 4685-4692). Institute of Electrical and Electronics Engineers Inc.. URL: https://doi.org/10.1109/ICRA.2018.8461160 DOI: https://doi.org/10.1109/ICRA.2018.8461160
10. Iqbal, A., Busso, C., Gans, N.R. (2015) Adjacent Vehicle Collision Warning System using Image Sensor and Inertial Measurement Unit. In Proceedings of the 2015 ACM on International Conference on Multimodal Interaction—ICMI ’15, Seattle, WA, USA, 9–13 November 2015; ACM Press: New York, NY, USA, 2015; pp. 291–298. DOI: https://doi.org/10.1145/2818346.2820741
11. C. Chen, L. Liu, T. Qiu, Z. Ren, J. Hu, and F. Ti, “Driver’s intention identification and risk evaluation at intersections in the Internet of vehicles,” IEEE Internet Things J., vol. 5, no. 3, pp. 1575–1587, Jun. 2018. DOI: https://doi.org/10.1109/JIOT.2017.2788848
12. Halim Z., Sulaiman M., Waqas M. (2022). Deep neural network-based identification of driving risk utilizing driver dependent vehicle driving features: a scheme for critical infrastructure protection // Journal of Ambient Intelligence and Humanized Computing, 2022, 14(2):1-19. DOI:10.1007/s12652-022-03734-y DOI: https://doi.org/10.1007/s12652-022-03734-y
13. Oprea,S., Martinez-Gonzalez,P., Garcia-Garcia,A., Castro-Vargas,J.A., Orts-Escolano,S., Garcia-Rodriguez,J., Argyros,A. (2020). A review on deep earning techniques for video prediction / IEEE Transactions on Pattern Analysis and Machine Intelligence.
14. Kushwaha, M.; Abirami, M.S. (2023). Intelligent Model for Avoiding Road Accidents Using Artificial Neural Network, International Journal of Computers Communications & Control, 18(5), 5317, 2023. https://doi.org/10.15837/ijccc.2023.5.5317 DOI: https://doi.org/10.15837/ijccc.2023.5.5317
15. Maaloul, B. (2018). Video-based algorithms for accident detections. Ph.D. thesis. Universit. Ede Valencienne set du HainautCambresis; Universit. edeMons.
16. Robles-Serrano, S., Sanchez-Torres,G., Branch-Bedoya,J., 2021. Automatic detection of traffic accidents from video using deep learning techniques. Computers 10, 148. DOI: https://doi.org/10.3390/computers10110148
17. Coll-Josifov, R., Masip-.Alvarez, A., Lav`ernia-Ferer, D., 2022. Deep learning classification applied to traffic accident sprediction, in: XLIII Jornadasde Autom.atica, Universidadeda Coru˜na .Servizode Publicaci. ons. pp. 964–971. DOI: https://doi.org/10.17979/spudc.9788497498418.0964
18. Сергиенко Н.Е., Маренич А.Н. Бортова система контролю стану водія // Автомобільний транспорт, 2010. Вип. 27. С. 148-152.
19. Purkrábková, Z., Růžička, J., Bělinová, Z., Korec, V. Traffic accident risk classification using neural networks // Neural Network World, 2021, Issue 5, p.343 DOI: https://doi.org/10.14311/NNW.2021.31.019
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
