Study On Method Of Driver Recognition Under The Condition Of Vehicle Networking

With the rapid development of social economy and motorization,road traffic safety and automobile safety performance have become an urgent problem to be solved in the harmonious development of cities.With the development of wireless communication technology,sensor technology,and the arrival of big data era,currently vehicle networking has become the key to the development of intelligent transportation systems and the improvement of road traffic safety level.Driving behavior is the main factor that causes traffic accidents and road congestion,driving behavior identification in vehicle networking is an important method to predict and reduce road traffic accidents.Therefore,this paper mainly discusses the method of driver behavior recognition in vehicle networking.Firstly,by comparing the traffic simulation model and simulation software,VISSIM is used to establish the urban and high-speed road network environment.By analyzing and setting the parameters of VISSIM driver module,the local road network simulation under different driving conditions is achieved and massive basic simulation data is obtained.Secondly,the transformation from basic driving parameters to running status parameters is realized,which is based on the simulation data of VISSIM and combined with the theory of vehicle dynamics.After that,aiming at the deficiency of the classical rough set,this paper proposes neighborhood rough set for feature reduction,which improves the operation speed of subsequent analysis.The sample data after attributes reduction is desposed by the method which is combinated with ensemble empirical mode decomposition(EEMD),correlation coefficient and sample entropy.Then the obtained sample entropy value is used as the input feature vector of the clustering,which makes up for the disadvantages of applying statistical methods to deal with driver behavior data.Finally,as different driver behavior recognition method,Gath-Geva fuzzy clustering algorithm is used to construct the driver behavior recognition model,which takes the cluster centers as the standard vector.By means of the minimum average closeness principle,different driver behavior could be achieved.According to the clustering results to improve the identification model,the results show that the driver behavior recognition has made good effect.