Driver's decision-making for merging behavior at freeway merging section

Abstract

A merging section is where two different flows conflict due to active bottlenecks. A merging vehicle attempts to change lanes safely and in as short a time as possible without risking collision with the main lane vehicle. Main lane vehicle keeps its initial car-following status and tries to minimize change in driving behavior caused by interfering vehicles. Conflicts that occur between two vehicles traveling in a merging section cause problems at local points on the freeway and cause upstream congestion. In this paper, based on game theory, decision-making model is proposed that considers interactions about joining decisions of merging vehicles and yielding decisions of main lane vehicles. This model was developed in terms of travel time it takes to pass through a merging section. Numerical study using the Nash equilibrium was conducted to distinguish driver decision-making, and vehicle trajectory data was used to verify the proposed model. As a result of the model verification, final results predicting driver decisions with accuracy of 90% or more were derived. Through a microscopic analysis and decision-making study of merging behavior, it is possible to understand clearly the lane changing interaction process of two vehicles. This provides the basis for the traffic management strategy of freeway merging sections.