Optimization And Algorithm Design Of Road Forecasting System For Incompleye Telematics Environment

With the continuous expansion of China’s urban scale and development of urbanization,the congestion caused by the gathering of vehicles in urban areas is intensified.Internet of vehicle or Telematics,alleviate congestion by feedback road information real-timely.However,at present and for a long time to come,the coverage rate of Telematics is still at a low level,called as Incomplete Telematics Environment(ITE).Base on the limited data of ITE,this paper studies effectively uses the data collected by Telematics to optimize and design the road prediction system to predict the urban traffic conditions accurately,alleviate urban congestion effectively.The main contribution of this thesis is summarized as follows:(1)Based on the characteristics of vehicle distribution and data sharing in ITE,a real-time distributed road prediction system for ITE is proposed.The system contains the design and optimization for three parts of Telematics environment fusion distributed framework platform,path planning algorithm and user interaction interface,laying the foundation of using multi-dimensional road data to predict road conditions.(2)Innovatively merge three-time dimensions data contain with expected navigation data information,real-time GPS information and historical road condition information to the automatically updating path planning algorithm.Design the method of extracting navigation path according to the road condition data,and combines the distributed road prediction system to efficiently share surrounding vehicle driving data and road condition real-timely to achieve accurate prediction of further road data.(3)Build a Web interaction page with the front-end page service framework Flask as an interface between users and the road prediction system.Through the front-end page,the road conditions and navigation path information predicted by the system are displayed visually.From back-end to front-end and data source to analysis result,the road prediction function of real-time data stream in ITE is designed and optimized.Finally,using the open-source data,we design road experiments in different environments.The average speed of the city is increased by 18%,87.9%of the connected vehicles reach their destinations with shorter driving time.Effectiveness and reliability of the system is proven by experiments.