Research And Implementation Of Cooperative Map Matching Based On Internet Of Vehicles

In recent years,with the improvement of economic and living standards,there has been an explosive growth of private cars in China.And there is a series of traffic problems such as traffic congestion,frequent accidents,air pollution because of the rapid growth of vehicles.The Internet of Vehicles and automatic driving technology,as the hybrid technology of information,communication and control,have become one of the important ways to solve urban traffic problems.However,the Internet of Vehicles and automatic driving technology put forward higher requirements for vehicle positioning accuracy,while the traditional Global Navigation Satellite System(GNSS)receiver can't achieve lane level positioning.Based on the platform of the Internet of Vehicles,a new algorithm is proposed to improve the accuracy of vehicle positioning by using the real-time positioning data and road constraints of other connected vehicles in this paper.A new method is also provided to meet the needs of high-precision positioning in the environment of the Internet of Vehicles.The main contents of this paper are as follows:(1)A cooperative map matching algorithm based on adaptive genetic Rao-Blackwellized particle filter is proposed.The algorithm relies on the platform of the Internet of Vehicles to share the real-time positioning information and improve the accuracy of vehicle positioning.At the same time,the algorithm is simulated with the software of Prescan and Matlab.The simulation results show that the algorithm has a great improvement in positioning performance compared with the traditional positioning algorithm.And the estimation accuracy of the new algorithm is higher.(2)The adaptability and robustness of the cooperative map matching algorithm proposed in this paper are deduced by some theory models and verified by simulation experiments.The results show that the positioning accuracy of GNSS receiver under different road conditions and different number of connected vehicles are improved.In addition,with the increase of the number of connected vehicles,the accuracy of the estimation of common deviation will be improved.And the accuracy of the vehicle positioning will also be improved.Apart from this,the positioning accuracy of this algorithm is also affected by the communication between vehicles and the accuracy of digital map.In this paper,the Dedicated Short Range Communication(DSRC)mode is selected in the actual test.However,with the application of C-V2 X technology,the development of high-precision map and the improvement of relevant geographic information security laws,the positioning accuracy of traditional GNSS receiver will be improved more.(3)The intelligent vehicle terminal based on DSRC is used to share the information of real-time positioning and map road constraint in actual test.Then,the collected data is processed by the cooperative map matching algorithm to get the positioning results.Finally,the positioning results of real time kinematic(RTK)device are taken as the reference data.And the positioning results are compared with the traditional GNSS receiver and single map matching.The experimental results show that with four connected vehicles,the range of positioning error of cooperative map matching is 1.52 m.It is only 39.33% and 54.92% of the traditional GNSS and single vehicle map matching positioning results.At the same time,the positioning accuracy(CEP)of this algorithm is 0.97 m,which is 2.55 m higher than the raw GNSS positioning result.The experimental results show that there is a better positioning result of this algorithm.