A method for analyzing interaction of driver intention through vehicle behavior when merging

Estimation of driver intention is a key technology in the advancement of driver assistance systems. Previous studies in regard to driver intention estimation of lane change primarily involve situations where there is no great deal of mutual interaction between the driver attempting to change lanes and the drivers of other vehicles. In contrast, this study looks at situations where there is a mutual interaction between drivers, such as the scenario when vehicles merge because of reduction in number of lanes on city roads. This paper addresses the problem of the driver intention estimation in this type of situation by proposing a novel approach to analyze the interaction of driver intent through the vehicle behavior. We first present the method, and then, show the results of applying the method to analyze the interaction between the drivers of two vehicles in neighboring lanes when merging in an experiment on a driving simulator.