DEVELOPMENT OF A MODERN MODEL FOR PREVENTING ROAD ACCIDENTS USING A CONVOLUTIONAL NEURAL NETWORK
DOI:
https://doi.org/10.14308/ite000769Keywords:
data analysis, convolutional neural network, road safety, road parameters, road accident prevention, PythonAbstract
The article examines the possibilities of artificial neural networks for avoiding traffic accidents by motor vehicle drivers. The testing was carried out in the settlement. Ensuring road traffic safety (hereinafter ̶ BDR) is a component of national tasks for ensuring personal safety, solving demographic, social and economic problems, as well as improving the quality of life and promoting the development of cities and villages.
Based on a generalized analysis of existing state programs to avoid road accidents, it can be noted that the prospect of creating new methods is consistent with the priorities and programs of Ukraine's socio-economic development in the long term and is in the field of global road safety. After analyzing similar software products, it was found that the number of foreign and Ukrainian companies engaged in the development of communication networks for road transport with different architectures is constantly growing.
The paper recommends the use of a systematic set of measures due to an integrated software approach, and a competent analysis of the results obtained shows the development of standards and performance indicators for use in the field of road safety. The use of modern technologies for road users to support road safety, the organization of driver training at the level and the international exchange of experience play an important role in overcoming these problems.
The application of the method of building a convolutional neural network to prevent road accidents described in the study can be used in the activities of both government agencies, such as the State Special Transport Service, the Patrol Police of Ukraine, units of the National Police of Ukraine and other central bodies of state executive power, enterprises, their associations, institutions and organizations.
Downloads
Metrics
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.