Prediction of traffic accidents by using neural network

• Autorzy: Růžička J. , Bělinová Z. , Korec V. , Cikhardtová K. , Tichý T.
• Języki publikacji: EN

Abstrakty

EN

The prediction of traffic accidents in urban networks is one of the key future theme in the areas of traffic control and navigation. Early identification of the risk of a traffic accident can lead to an increase in the safety and smoothness of road transport. Neural networks belong to the expert methods for modeling of complex systems. The issue of their use in the transport sector is scientifically quite progressive. The article describes the design of prediction model based on available traffic data from town Uherské Hradiště. Traffic data was collected from many sources, e.g. junction detectors, meteorological stations or traffic accident portal. Appropriate parameters for the model were selected from the traffic data. The model was then tested on a 2-month data sample. The aim of the article is to confirm the suitability of using neural networks to predict traffic accidents.

Słowa kluczowe

EN

traffic accident; prediction; neural network

PL

wypadek drogowy; prognoza; sieć neuronowa

• Wydawca: Polskie Stowarzyszenie Telematyki Transportu
• Czasopismo: Archives of Transport System Telematics
• Rocznik: 2019
• Tom: Vol. 12, iss. 4
• Strony: 22--26
• Opis fizyczny: Bibliogr. 10 poz.

Twórcy

autor

Růžička J.

• CTU IN PRAGUE, Konviktská 20, 110 00 Prague, Czech Republic

autor

Bělinová Z.

• CTU IN PRAGUE, Konviktská 20, 110 00 Prague, Czech Republic

autor

Korec V.

• CTU IN PRAGUE, Konviktská 20, 110 00 Prague, Czech Republic

autor

Cikhardtová K.

• CTU IN PRAGUE, Konviktská 20, 110 00 Prague, Czech Republic

autor

Tichý T.

• CTU IN PRAGUE, Konviktská 20, 110 00 Prague, Czech Republic

Bibliografia

• [1] BISKUP R.: Možnosti neuronových sítí, disertační práce, Česká zemědělská univerzita, Praha, 2009
• [2] OBITKO M.: Interactive tutorials, mainly from the area of artificial intelligence. [online], 1999 [date of access: 22.01. 2019]. Dostupné z URL: <http://www.obitko.com/tutorials/>.
• [3] SAMEEN M., PRADHAN B.: Severity Prediction of Traffic Accidents with Recurrent Neural Networks. Applied Sciences [online], 7(6) [date of access: 22.01.2019]. DOI: 10.3390/app7060476. ISSN 2076-3417. Dostupné z: http://www.mdpi.com/2076-3417/7/6/476, 2017
• [4] ALKHEDER S., TAAMNEH M., TAAMNEH S.: Severity Prediction of Traffic Accident Using an Artificial Neural Network. Journal of Forecasting [online], 36(1), 100-108 [date of access: 22.01.2019]. DOI: 10.1002/for.2425. ISSN 02776693. Dostupné z: http://doi.wiley.com/10.1002/for.2425, 2017
• [5] JABAR Y., RABABAA M.: Neural Technique for Predicting Traffic Accidents in Jordan. journal of american science. 9. 347-358, 2013
• [6] RŮŽIČKA J., et al.: Implementace systémů řízení a senzorických sítí ve městě; Silniční obzor, 2018(5), 138-141. ISSN 0322-7154, 2018
• [7] RUZICKA J., et al.: Methods of traffic surveys in cities for comparison of traffic control systems - A case study. In: 2018 Smart City Symposium Prague (SCSP) [online]. IEEE, 2018, s. 1-6 [date of access: 22.01.2019]. DOI: 10.1109/SCSP.2018.8402666. ISBN 978-1-5386-5017-2. Dostupné z: https://ieeexplore.ieee.org/document/8402666/, 2018
• [8] BOYARKIN I., et al.: Implementation of new adaptive control algorithms in the defined urban area. In: 2018 19th International Carpathian Control Conference (ICCC) [online]. IEEE, 2018 s. 608-612 [date of access: 22.01.2019]. DOI: 10.1109/CarpathianCC.2018.8399702. ISBN 978-1-5386-4762-2. Dostupné z: https://ieeexplore.ieee.org/document/8399702/, 2018
• [9] GHAFARI E., et al.: Full Paper-BMC 2012-EG-04-1 revEJ, 2014
• [10] mapy.cz. [Online] [date of access: 2.10.2018] https://mapy.cz.

Uwagi

PL

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).