Assessing and Modeling Speed of Vehicles in Arterial Two-way Tunnels

The way drivers behave on roads is based on the state of the environment around them. Previous research shows that the vehicle starts to slow down at a certain distance from the entrance of tunnel. The main purpose of this research is to model the speed of vehicles entering the urban tunnels based on their velocity changes before entering the tunnel using neural-fuzzy network. 30 different drivers, were examined in similar conditions to study the behaviors of drivers. The study was conducted in a Renault Logan with manual transmission. Using Pearson correlation analysis, the relationship between the input speed variables to the tunnel and the vehicle speed variations is investigated.. The correlation coefficient value is -0.7 which means that there is a negative correlation between the two variables. The results show that the neural-fuzzy neural network method is able to predict velocity changes based on the lateral distance of the cars with the tunnel wall.. The results of this study are used to analyze the behavior of drivers in suburban tunnels.. Given the importance of abrupt changes in speed and transverse movement of cars, especially on two-way lanes, it can be possible to increase tunnel safety by reducing stressors.