Research On Real-Time Traffic Flow Guidance System For Internet Of Vehicles

In recent years,due to the increase in the number of vehicles and the growth of traffic network but the lag of traffic information service facilities,traffic safety problems and urban congestion problems are outstanding.In addition,traffic congestion has caused more energy consumption and exhaust emissions of vehicles.People increasingly hope that traffic can be more intelligent,and the service can be more convenient.Therefore,as an important solution of intelligent transportation system,the internet of vehicles has been widely concerned by the community.Based on real-time traffic information in the internet of vehicles,to study and design more convenient and humanized traffic flow guidance system is the core content of intelligent transportation system.In the internet of vehicles,the vehicles' real-time information(latitude and longitude,speed,direction,etc.)will be periodically uploaded to the cloud server.According to all vehicles' real-time information in the cloud server,we can calculate the real-time traffic conditions,and then forecast the future short-term traffic conditions,and finally,based on real-time and forecasting traffic conditions,find an optimal travel path.In the vehicle navigation process,the travel route will be optimized real-time dynamically.Based on the vehicle network,this paper studies and design the feasible and efficient traffic flow guidance system with real-time traffic information in the internet of vehicles and finally verify the feasibility and effectiveness of the system by simulation.The main research work is as follows:1.According to all vehicles' status information uploaded real time from the internet of vehicles,we calculate the real time traffic flow of all roads,which is more real-time and accurate than any of the existing traffic flow parameter acquisition methods.2.In the case of giant database of traffic history data,KNN algorithm is implemented on the Spark calculation platform with real-time traffic flow information,and the feature vector with time-space characteristics is adopted in the KNN algorithm.The algorithm have high computational efficiency,accurate prediction results,and strong sense of perception.3.Based on the real-time and predicted traffic flow information,the weighted average method is used to calculate the travel time of each section of the road.Then,the path planning is carried out by using the algorithm of contraction hierarchical,and the optimal route is optimized dynamically according to the road traffic condition.4.We simulate the real-time traffic flow guidance system using California's road map and the traffic data of the California PeMS system.The simulation results show that the system can effectively utilize the real-time and predicted traffic flow data,and provide the time-saving and reliable path planning results for the car driver,and dynamically optimize the path during vehicle travel.