格网建模辅助的RTK部分模糊度固定算法  

Grid modeling aided partial ambiguity resolution algorithm for RTK

作　　者：赵彩霞 孙蕊 蒋磊 盛琪 钟德政 ZHAO Caixia;SUN Rui;JIANG Lei;SHENG Qi;ZHONG Dezheng(Nanhang Jincheng College,Nanjing 211156,China;College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)

机构地区：[1]南京航空航天大学金城学院,南京211156 [2]南京航空航天大学民航学院,南京211106

出　　处：《导航定位学报》2025年第1期39-51,共13页Journal of Navigation and Positioning

基　　金：国家自然科学基金项目(42222401,42174025,41974033);江苏省自然科学基金项目(BK20211569)。

摘　　要：针对传统部分模糊度固定(PAR)方法在复杂城市区域中难以准确评估全球卫星导航系统(GNSS)观测信息的质量,导致单历元实时动态差分定位(RTK)的模糊度固定率和定位精度通常较低的问题,提出一种格网建模辅助的部分模糊度固定算法,以提升单历元RTK在城市环境中的精度与可用性:通过多日连续采集相同城市区域内的GNSS先验信息,离线构建一种基于集成装袋树(EBT)的伪距误差格网预测模型,从而获得该格网模型覆盖范围内各观测卫星的伪距误差预测规则;然后,基于用上述规则在线预测得到的伪距误差结果,提出一种格网建模辅助的部分模糊度子集确定方法,用于提升单历元RTK的模糊度固定率。实验结果表明,提出算法相较于传统的全模糊度固定算法和基于卫星高度角辅助的PAR算法,能显著提高单历元RTK在复杂城市区域中的模糊度固定成功率,从传统算法的37.37%和64.95%提高到85.24%,RTK三维定位精度的提升率则分别达到43.56%和32.14%。Aiming at the problem that it is difficult for traditional partial ambiguity resolution(PAR)methods in accurately assessing the quality of global navigation satellite system(GNSS)observations in complex urban areas,which often leads to lower ambiguity fixing rate and positioning accuracy in single epoch real-time kinematic(RTK)applications,the paper proposed a grid modeling aided PAR algorithm to enhance the accuracy and availability of single-epoch RTK positioning in urban environments:an offline pseudorange error grid prediction model based on ensemble bagging tree(EBT)was constructed by collecting prior GNSS information in the same area over several days,in order to obtain the pseudorange error prediction rule for all GNSS satellites in the area;then,based on these online predicted pseudorange error results by using the above rule,a grid modeling aided partial ambiguity subset selection strategy was put forward to enhance the ambiguity fixed rate of the single-epoch RTK.Experimental result showed that compared to the traditional full ambiguity resolution algorithm and the satellite elevation angleassisted PAR algorithm,the proposed method could help significantly improve the success rate of ambiguity resolution for singleepoch RTK in urban areas,increasing from 37.37%and 64.95%to 85.24%,respectively,and the improvements in RTK threedimension(3D)positioning accuracy would be 43.56%and 32.14%,respectively.

关 键 词：全球卫星导航系统(GNSS) 实时动态差分定位(RTK) 格网建模 伪距误差预测 部分模糊度固定(PAR) 

分 类 号：P228[天文地球—大地测量学与测量工程]