The merging behavior in urban expressway weaving section affects the operation efficiency of the weaving area and reduce the efficiency of expressway traffic. This article obtains the video data, selects the merging position and speed of the merging vehicle for analysis. Clustering analysis method is used to evaluate the vehicle's merging position. The contour coefficient is used for evaluate the results. Speed analysis result shows that the optimal cluster number is 3. 89.04% of the vehicle speeds are within the ±2.5mꞏs-1 interval of the main line speed, 42.5% of vehicles merged at a speed lower than the main line speed, and 57.5% of vehicles had a speed higher than the main line traffic speed. According to the above analysis, the merging behavior of expressway interweaving area is classified. And choose the support vector machine algorithm to predict it. The results show that the accuracy of the SVM model is above 78%, indicating that the SVM model can better predict the merging behavior of vehicles in the interweaving area on the ramp.
Characteristic analysis and behavior prediction method of converging behavior in urban expressway weaving section
International Conference on Smart Transportation and City Engineering 2021 ; 2021 ; Chongqing,China
Proc. SPIE ; 12050
2021-11-10
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Real-time crash prediction for expressway weaving segments
Elsevier | 2015
|Real-time crash prediction for expressway weaving segments
Online Contents | 2015
|Characteristic Parameters Identification of Macro Traffic Flow on Urban Expressway
Trans Tech Publications | 2013
|