In order to objectively describe the operation characteristics of vehicles on bad weather congested roads, this paper selects representative urban roads for data collection. The experimental data were preprocessed, and the vehicle driving data were divided into kinematic segments. Then principal component analysis and K-means ++ clustering were carried out to screen out the kinematic segments that could represent the actual driving of vehicles, and the vehicle driving conditions of urban congested sections under severe weather were obtained by connecting them.  Finally, the characteristic values of the constructed driving conditions are compared and verified with those of the actual driving data of the vehicle. The constructed driving conditions can truly reflect the actual running conditions of the vehicle.

Kinematics segment;Characteristic parameter value;PCA(principal component analysis);Driving condition