Lane change behavior on motorways based on naturalistic trajectory data

Abstract

Lane change behavior is a major aspect in traffic flow modeling. Since only a few empirical analyzes are available, validation of lane change models is frequently limited to macroscopic characteristics (e.g. number of lane changes). In addition, many existing lane change models do not reflect lane change behavior in complex situations realistic enough. Therefore, we investigate naturalistic trajectory data from German motorways and analyze gap acceptance behavior focusing on varying discretionary lane change objectives, especially on cooperative lane changes. We propose a methodology to classify different lane change objectives and analyze the critical time gap using the Raff´s method. The results show differences in gap acceptance between varying discretionary lane changes classes. Moreover, we found that drivers who perform a cooperative lane change accept rather low time gaps. The analyzes should provide a basis for validating existing and developing new lane change models.