Research On Application Of Data From Internet Of Vehicles Facing Intelligent Transportation

Along with the society and economy developing,the number of private cars is increasing,which results in many social issues,such as traffic jams,smog,deterioration of the global energy crisis and so on.In order to solve those problems,Intelligent Transportation System(ITS)was born.By integrating information technology,data transmission technology,electronic sensors technology,ITS enables traffic management real-time,accurate and efficient.An important development direction of ITS is Internet of Vehicles(IoV).IoV is a giant interactive network composed of vehicles location,velocity and routes.Through using central server to do analysis of data from vehicles' information network,IoV is able to forecast congestion and do real-time dynamic navigation for all cars,which can improve the efficiency of transport.This paper takes short-time traffic forecasting as the main theme of the research.Through the analysis of existing forecasting models and research on parameters optimization of support vector regression model,this paper proposes support vector regression(SVR)traffic flow forecasting model based on Genetic Algorithm and Particle Swarm Optimization(GAPSO).For the choice of parameters of SVR has a great impact on the accuracy of forecasting,and traffic data has unique characteristics,so this paper combines Genetic Algorithm(GA)and Particle Swarm Optimization(PSO)to design a new algorithm,which applies genetic selection,crossover and mutation from GA to the evolution of particle and uses adaptive search mechanism in the algorithm's parameters' updating,called GAPSO.Finally,this paper does experimental validation.This paper uses Mean Absolute Error(MAE),Mean Absolute Percentage Error(MAPE)and the Root Mean Square Error(RMSE)as the performance criterions.Experimental results show that GAPSO-SVR traffic flow model does accurate forecasting and has good convergence,which meets short-term traffic flow forecasting's real-time,accuracy,and reliability requirements.