Analysis of Lane Changing Behavior of Vehicles on Urban Roads Based on Similarity Measurement of Velocity Curves

Le résumé fourni par la source

In order to explore the impact of lane-changing behavior of a lane-changing vehicle on the following vehicle during free lane-changing on urban roads, the trajectory data of the two vehicles were extracted from the video recorded by a video recorder. The Euclidean distance and DTW distance were used to measure the similarity of the speed trajectories of the two vehicles. Manual scoring was used as the actual situation of the speed trajectory similarity. The accuracy of the results of the two methods was compared, and the variables affecting the trajectory similarity were determined through correlation analysis. The research results show that: DTW is more accurate than Euclidean distance in expressing the similarity of speed trajectories; and when the longitudinal distance between the two vehicles at the time of lane-changing intention and the time of lane-changing is small, DTW is small; the smaller the speed difference at the beginning of lane-changing, the smaller the DTW; the DTW of the lane-changing type entering the fork is significantly greater than that of other lane-changing types; DTW is larger when there is a car in front of the lane-changing vehicle and a car behind it.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.