Research on the Lane-Changing Behavior of Drivers at Highway Interchanges

To delve into the characteristics of drivers’ lane-changing behavior within highway interchanges, this study firstly used the Tracker tool to extract vehicle lane-changing trajectory data from videos. Subsequently, a random forest regression model was employed to analyze the most crucial influencing factors of vehicle lane-changing. At last, traditional indicators, TTC and DRAC, along with an accident tree’s risk categorization standards, were used to classify the lane-changing risk. The research results show that lane-changing behavior is most profoundly influenced by the longitudinal spacing with the preceding vehicle after the lane change, with an importance of 21.2%. The refined accident tree safety risk categorization method proves to be more accurate. The results of this study can serve as a methodological reference for drivers’ lane-changing behaviors on highways and offer rational recommendations for future traffic safety management.