顾及通信延迟的车路协同高精度定位  

Vehicle high-precision positioning considering communication delay for intelligent vehicle-infrastructure cooperation system

作　　者：张红娟 钱闯[3] 招倩莹 李文卓 李必军 ZHANG Hongjuan;QIAN Chuang;ZHAO Qianying;LI Wenzhuo;LI Bijun(State Key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,Wuhan 430079,China;Engineering Research Center for Spatio-temporal Data Smart Acquisition and Application,Ministry of Education of China,Wuhan University,Wuhan 430079,China;Intelligent Transportation Systems Research Center,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]武汉大学测绘遥感信息工程国家重点实验室,湖北武汉430079 [2]武汉大学时空数据智能获取技术与应用教育部工程研究中心,湖北武汉430079 [3]武汉理工大学智能交通系统研究中心,湖北武汉430063

出　　处：《测绘学报》2024年第1期101-117,共17页Acta Geodaetica et Cartographica Sinica

基　　金：国家重点研发计划(2021YFB2501100);武汉理工大学自主创新研究基金(223144001)。

摘　　要：近几年,随着智能交通和通信技术的发展,智能车路协同系统引起了广泛的关注。车辆的位置特征是智能交通中的基本元素。车路协同环境下,车辆可通过通信设备接收路侧端的定位信息进行自车定位。本文旨在解决车路协同环境中不稳定的通信延迟带来的定位误差的问题,提出了一种基于因子图的考虑通信延迟的车辆高精度定位模型。在GNSS的情况下,基于路侧激光雷达点云LiDAR聚类方法识别和定位目标车辆,通过4G通信网络将目标定位结果发送至车辆,采用因子图将当前时刻的车载惯性测量单元IMU测量信息与滞后的路侧目标定位结果直接融合,基于增量平滑推理方法,实现车辆位置、速度和姿态的最优估计。最后,结合实测和仿真数据,利用实车试验验证了本文方法,与传统处理时间延迟的外推法对比分析,结果表明本文方法可提高车辆的定位和测速精度,并消除了高度不稳定的通信延迟对定位的影响。In recent years,with the development of intelligent transportation and communication technology,intelligent vehicle-infrastructure cooperation systems have attracted widespread attention.The location features of vehicles are the basic elements in intelligent transportation.In the vehicle-infrastructure collaborative environment,the vehicle can receive the positioning information of the roadside unit through the communication device for self-vehicle positioning.This paper aims to solve the problem of positioning errors caused by unstable communication delays in the vehicle-infrastructure collaborative environment and proposes a high-precision vehicle positioning model based on factor graphs that considers communication delays.In the absence of global navigation satellite system(GNSS)information,the target vehicle is identified and located based on the roadside light detection and ranging(LiDAR)point cloud clustering method.The target positioning result is sent to the vehicle through the 4G communication network.The factor graph is used to directly fuse the measurement information of the vehicle inertial measurement unit(IMU)at the current moment with the lagging roadside target location results.Based on the incremental smoothing inference method,the optimal estimation of the vehicle position,speed and attitude is realized.Finally,combined with the measured and simulated data,the method proposed in this paper is verified by real vehicle experiments.Compared with the traditional extrapolation method of processing time delay,the results show that our method can improve the accuracy of vehicle positioning and speed measurement and eliminate the influence of highly unstable communication delay on positioning.

关 键 词：车路协同 通信延迟 高精度定位 因子图 路侧感知 目标识别与定位 

分 类 号：P227[天文地球—大地测量学与测量工程]