The Study Of Driving Tendency Characteristics Of Entrance And Exit Of Urban Expressway

The entrance and exit of urban expressway is the bottleneck area of expressway operation,and the vehicles’ diverging,merging and interweaving operations occur frequently,which is very easy to cause traffic conflicts and accidents.In the induced factors of traffic accidents,the proportion of human factors is more than 90%.The differences among drivers lead to different driving tendency characteristics in the driving process.From the point of view of considering the individual differences of drivers,the driving tendency characteristics reflecting the individual differences of drivers are introduced into the study of traffic safety at the entrances and exits of expressways,so that personalized active safety protection can be implemented for different types of drivers,which is of great significance to improve the safety level of Expressway entrances and exits.In this paper,the characteristics of driving tendency at the entrance and exit of urban expressway are studied.First of all,by designing a questionnaire for the entrance and exit of the expressway,this paper collected the data including the entrance and exit driving behavior,driving attitude,driving technology and personality characteristics.Based on factor analysis and K-means clustering,the drivers were divided into three kinds of static driving tendencies.Combined with the scores of the questionnaire,the static driving tendencies were divided into conservative,general and radical types Based on the accident experience,the results of driving tendency classification are verified and analyzed.Secondly,based on the real traffic observation at the entrance and exit of the expressway,this paper constructs the driving simulation scene at the entrance and exit of the expressway by using the eight degree of freedom driving simulation system,selects 15 drivers from each of the three types of drivers for driving simulation experiments,extracts the dynamic driving characteristics of different types of drivers in the process of diverging,and extracts the dynamic driving characteristics from the two aspects of vehicle operation state characteristics and driver operation characteristics This paper analyzes the dynamic driving characteristics of drivers with different driving tendencies.Generally speaking,with the increase of radicality,the speed level of drivers tends to increase,while the stability level of vehicles tends to decrease.Based on the results of Kruskal Wallis nonparametric test,it shows that the indicators have obvious differences among different driving tendency groups.Finally,based on the 12 driving characteristic index data selected by dynamic driving characteristic analysis,the dimension of the original data is reduced by principal component analysis,five principal components are extracted as the input layer of neural network,and the static driving tendency data is taken as the output layer of neural network.A driving tendency identification model based on principal component analysis and BP neural network is constructed.The results show that the model is effective for driving.The recognition rate of propensity type is more than 86%,which can adapt to the dynamic recognition of driving propensity type of drivers at the entrance and exit of expressway.