Modeling of speed in vehicles entering two-way suburban tunnels by Adaptive Neuro Fuzzy Inference System

Abstract

The behavior of drivers on the roads is elicited from the state of the surrounding environment. The author's research shows that the vehicle starts to decelerate at a certain distance from the tunnel when it is observed, and they have the lowest speed when reaching the beginning of the tunnel. As soon as the tunnel is passed, the vehicle increases speed again in a certain length. The main purpose of this study is to model the speed of vehicles entering suburban tunnels based on the speed changes before entering the tunnel using the neuro-fuzzy network. Then, to validate the designed model, the data of 30 different drivers were used who travel in the same conditions by a Renault Logan vehicle with a manual transmission system. Using the Pearson correlation analysis method, the relationship between the variables of the speed of entrance to tunnel and changes in vehicle speed was investigated. The value of the correlation coefficient is equal to -0.7, which means the strong negative correlation between the two variables. The results show that the neuro-fuzzy network method has the ability to predict speed changes with a high accuracy based on the initial speed of entrance to the tunnel. The results of this study are used to analyze the behavior of drivers in suburban tunnels. Due to the importance of abrupt speed changes in an unusual way, especially on two-way routes, the safety of tunnels can be increased by reducing the stressors in drivers.