@article{ref1,
title="Neuro-fuzzy cooperative collision warning system in based on driver behavior in chain accident using connected vehicles",
journal="Journal of transportation research (Tehran)",
year="2023",
author="Eftekhari, Hamidreza",
volume="20",
number="2",
pages="339-352",
abstract="Failure to pay attention to the difference in drivers' perceptions of danger in collision warning systems increases unnecessary warnings. In this research, a model based on connected vehicle technology and based on driver behavior is presented. In the proposed model, the system is activated based on risk perception index. Situations that are potentially dangerous are then identified using a neural network. In the next step, the brake acceleration difference is calculated using an adaptive neural fuzzy estimator between two situations when the driver receives a warning or he does not. Finally, based on the driver's brake history, he will be warned. The results are evaluated on NGSIM benchmark data set, which contains more than 11 million records of driving behavior of about 3300 drivers, using MATLAB software. The accuracy of the system in detecting potential risk situations is 97%.The accuracy of the warning system is 98%. Accordingly, the proposed model increases the driver's confidence and customizes warning system based on driving behavior by eliminating unnecessary warnings.<p /> <p>Language: en</p>",
language="en",
issn="1735-3459",
doi="10.22034/tri.2022.294376.2928",
url="http://dx.doi.org/10.22034/tri.2022.294376.2928"
}