Research On Multi-GNSS Real-time Dynamic Deformation Monitoring Data Processing Method And Software Implementation

GNSS technology has been widely used in real-time deformation monitoring because it has the advantages of fast positioning,all-weather,automation,no need for viewing between stations and high precision.However,in the actual deformation monitoring,due to the complexity of the monitoring object,the single-system deformation monitoring often has the problem that the number of visible satellites is too small to complete the positioning,which affects the usability of GNSS technology in deformation monitoring;The gross error is unavoidable in measurement and the gross error will lead to a decrease in the accuracy of the monitoring results.In severe cases,it will even result in the divergence of the filter solution results,which will affect the reliability of the GNSS technology in deformation monitoring.In real-time deformation monitoring,influenced by different monitoring objects and the different monitoring environments in which the monitoring objects are located,the mathematical models are often not precise enough to describe the monitoring body movement state.The inaccuracy of the model affects the universality of the established deformation monitoring model.In view of the above problems,based on the derivation of a multi-GNSS real-time dynamic deformation monitoring algorithm,this paper studies the robust Kalman filtering algorithm and the adaptive Kalman filtering algorithm,and implements the above functions through software.Finally,the measured data are used to teste and analyze the functions of the software.The main content of this article are listed below:(1)The model for real-time deformation monitoring of GPS/BDS/GLONASS/Galileo multi-systems is studied.Starting from the basic observation,the original observation equation suitable for short-baseline real-time deformation monitoring,single-difference observation equation and the loose combination and tight combination model of double-difference observation equation were deduced respectively;the elevation-angle stochastic model and the SNR stochastic model were introduced.,and derived a stochastic model of double-difference observations;introduced common parameter estimation methods in deformation monitoring and deduced the basic formulas of Kalman filtering;introduced common ambiguity resolution methods,and deduced the ambiguity-based Search Technique formulas.(2)The relationship between adaptive robust Kalman filter and other estimation methods is studied;the common robust Kalman filter methods are summarized,and the robust equivalent weight formulas of the most widely used IGG III scheme are given.The common three types of adaptive Kalman filtering methods are introduced:the maximum post-estimation adaptive Kalman filtering,the adaptive Kalman filtering based on the principle of variance estimation,and the variance compensation adaptive Kalman filtering,and the advantages and disadvantages of these three kinds of methods are introduced.(3)Real-time dynamic deformation monitoring software for multi-GNSS is developed.The multi-system solution performance,robust performance and adaptive performance of the software are tested and analyzed based on the measured data.The results show that:the software can solve four system data and in the ultra-short baseline,its horizontal precision can reach cm level,the vertical precision can reach cm level;in the short baseline of 13km,its horizontal precision can reach 3cm,the vertical precision can reach within 6cm accuracy;the effect of software's robust function is remarkable,when the baseline length is about 300m,the monitoring precision of the self-developed software in the horizontal direction can reach 5mm,and that of the robust results can reach 3mm,and the vertical direction can reach within 6mm;the adaptive function works well and when it comes to movement,the adaptive Kalman filter can react quickly to the displacement and the precision is higher than the original Kalman filter.