Research On Cooperative Positioning Technology For Connected Vehicles Based On Multi-Source Information Fusion

With the rapid development of 5G and Internet of Vehicles technology,autonomous driving technology has become a key technology in the national intelligent vehicle innovation and development strategy and has attracted wide attention in today’s academic circles.The fundamental purpose of autonomous driving technology is to provide people with safer and more comfortable driving experience.However,there are various shortcomings in one-car autonomous driving technology,especially in terms of vehicle collaboration capabilities.Connected autonomous driving can effectively improve the synergy between car and road and Internet by connecting vehicles to the network,improve the safety of autonomous driving,and reduce the cost of autonomous vehicles.Localization is the most critical issue in autonomous driving,and it is also the focus of this article.In connected autonomous driving architecture,when positioning information is transmitted through the network,network delays will cause additional positioning errors,and ignoring this error will lead to safety hazards.In response to this problem,this paper proposes a Positioning Calibration Mechanism(PCM)for connected autonomous vehicles.PCM first calculates the network time delay,then calibrates the position information according to the vehicle motion state information,and finally saves the calibrated position information in the MEC server to reduce the positioning error caused by one-way time delay.This paper conducted experiments in FangShan and XuChang to verify the performance of the PCM algorithm.For example,under 100ms time delay,the average positioning error without PCM is 119.67cm,while the average positioning error with PCM is 3.03cm.Experimental results show that PCM can effectively reduce the positioning error caused by one-way time delay to centimeter level.In the technology of networked automatic driving,the high-precision positioning of the vehicle still relies on the satellite positioning system.How to solve the problem of high-precision positioning of vehicles without satellite signals is the key to expanding the application of autonomous driving scenarios.In this paper,a method for collaborative positioning of connected vehicles based on the confidence propagation algorithm is designed by using the multi-source information collected by autonomous vehicle.This method uses the mutual measurement between vehicles and the state measurement of the vehicle itself to construct a confidence propagation network,and then the current position of the vehicle is estimated step by step according to the order of message propagation.This paper conducts simulation experiments on the accuracy of the algorithm in three scenarios.The experimental results show that the method has high accuracy and stability under the current networked autopilot architecture,and has certain practical significance.