基于因子图的导航定位技术应用分析与思考  

Analysis and reflection on the navigation and positioning application based on factor graph

作　　者：潘献飞[1] 宁治文 王茂松 吴文启[1] 吴美平[1] PAN Xian-fei;NING Zhi-wen;WANG Mao-song;WU Wen-qi;WU Mei-ping(College of Intelligence Science and Technology,National University of Defense Technology,Changsha Hunan 410073,China)

机构地区：[1]国防科技大学智能科学学院,湖南长沙410073

出　　处：《控制理论与应用》2023年第12期2130-2141,共12页Control Theory & Applications

基　　金：国家自然科学基金项目(62203455)资助.

摘　　要：不同于传统滤波方法,因子图为传统的导航信息融合计算提供了另一种解决途径,成为近年来的研究热点.本文对因子图方法在导航定位领域的应用进行了梳理,分析了因子图模型的数学推导方法,在此基础上阐述了基于因子图的定位与多传感器组合导航方法,详细分析了因子图在同时定位与建图(SLAM),全球卫星导航系统(GNSS),协同定位、组合导航以及故障检测领域的应用现状,给出了相关模型,并指出了现有方法的局限性.最后,文章进行了总结与讨论,为因子图方法进一步的深入研究提供了一些建议.Different from the traditional filtering methods,factor graph provides another solution for the traditional navigation information fusion calculation,and has become a research hot-spot in recent years.This paper combs the application of factor graph method in the field of navigation and positioning.We analyze the mathematical derivation method of factor graph model,and on this basis,we describe the method of location and multi-sensor integrated navigation based on factor graph.Then we analyze the application status of factor graph in the fields of SLAM,GNSS,cooperative positioning,integrated navigation and fault detection in detail,give the relevant models,and point out the limitations of the existing methods.Finally,we make a summary and discussion,and provide some suggestions for the further study of factor graph method.

关 键 词：导航定位 因子图 概率推断 信息融合 SLAM 

分 类 号：TN96[电子电信—信号与信息处理] TN713[电子电信—信息与通信工程]