期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2007
卷号：XXXVI-5/C55
出版社：Copernicus Publications
摘要：A dynamic model is the usual modus operandi of a Mobile Mapping System. The model solution, after linearization and discretization, is achieved using the Weighted Least-Squares (WLS) approach, which results in one of the various Kalman filter algorithms. However, implicit in the formulation is that neither the observation equation matrices nor the transition matrices at any epoch contain random entries. As such an assumption cannot always be guaranteed, we here allow random observational errors to enter the respective matrices. We replace the WLS by the Total-Least-Squares (TLS) principle - with or without weights - and apply it to this novel Dynamic Errors-in-Variables (DEIV) model, which results in what we call Total Kalman Filter (TKF). It promises to offer more representative solutions to the dynamic models of Mobile Mapping Systems over existing versions of Kalman filtering
关键词：Kalman Filtering; Total Least Squares; Total Kalman Filtering; Mobile Mapping; Errors-in-Variables Models; ; GPS/INS Integration
