The method for automating the processing of test results for the effectiveness of ADAS (DCAS) systems
Abstract
An innovative method of automated processing of test results of the effectiveness of intelligent driver assistance systems (Advanced Driver-Assistance System – ADAS, Driver Control Assistance Systems – DCAS) is proposed. Two approaches have been developed to determine the moment when the target visual signal is triggered: an approach based on the analysis of brightness changes in specified areas of the frame and a method using convolutional neural networks. Examples of the technical implementation of these approaches are given. A comparison of approaches in terms of accuracy, processing speed, and resistance to environmental conditions has been performed. The efficiency of the proposed method is demonstrated by the example of processing the test results of the effectiveness of the front collision warning system. It has been found that the neural network approach provides greater accuracy in detecting the target image than the approach based on the analysis of brightness changes in specified areas of the frame.
About the Authors
Renat F. Altdinov
MADI
Russian Federation
Russian Federation
undergraduate
Sergey R. Kristalny
MADI
Russian Federation
Russian Federation
Candidate of Sciences (Technical), associate professor
References
1. Новые методы испытаний систем автоматического экстренного торможения и опыт их применения / А. М. Иванов, С. Р. Кристальный, Н. В. Попов [и др.] // Труды НГТУ им. Р.Е. Алексеева. – 2018. – № 2(121). – С. 146-155. – DOI 10.46960/1816-210X_2018_2_146. – EDN XSELUT.
2. Исследование процесса экстренного торможения с применением системы АБС и без неё / А. Е. Гончарук, Е. С. Красавин, В. Д. Сморчков, С. С. Шадрин // Автомобиль. Дорога. Инфраструктура. – 2021. – № 4(30). – EDN LSPNOQ.
3. Актуальные вопросы совершенствования технического зрения при использовании на автомобилях / А. Г. Тыняный, С. Р. Кристальный, П. А. Красавин [и др.] // Автомобиль. Дорога. Инфраструктура. – 2026. – № 1(47). – EDN HCEEXC.
4. Safety effectiveness of forward collision warning systems in the vehicle fleet: A driving simulation study / O. Olufowobi, J. Ivan, K. Wang, N. Eluru // Accident Analysis & Prevention. – 2025. – Vol. 218. – P. 108078. – DOI 10.1016/j.aap.2025.108078. – EDN EEOXZU.
5. Goodfellow, I. Deep Learning / I. Goodfellow, Y. Bengio, A. Courville. – Cambridge, MA: MIT Press, 2016. – 800 p. – ISBN 0262035618.
Review
Рецензент: В.В. Гаевский, д-р техн. наук, доцент, МАДИ
Views:
19
JATS XML
