Kinematic Precise Point Positioning Using Multi-constellation GNSS Observations

Precise point positioning technology is a kind of non-difference positioning technology which was developed in the late 1990s.It combines the advantages of standard single point positioning and wide area differential positioning technology.It is another revolution after RTK/CORS-RTK technology.After more than 20 years of development,the problems of GNSS-PPP theory and practice have been well solved and applied.At present,it is developing in the engineering stage,but there are still some problems in practical application,the positioning accuracy and convergence time of PPP technology are limited by the observation conditions.Therefore,multi-system GNSS satellite combined dynamic PPP is a problem that must be studied in engineering practice.In order to combine the dynamic PPP theme of multi-system GNSS,the author deduces the multi-system combined dynamic PPP algorithm from UofC function model.Finally,the algorithm is verified by static imitation dynamic experiment and measured dynamic vehicle experiment.The main research work and conclusions are as follows:(1)The theoretical basis of precise point positioning technology is introduced.The PPP function model and error processing method are given.Finally,the PPP parameter estimation method is introduced.(2)Based on the PPP function model,the UofC model of multi-system combined PPP is deduced by considering the differences between systems.The state equations of random walk and constant acceleration(CA)models are studied and the Random model of PPP estimation is constructed.Finally,based on MATLAB,the algorithm of multi-system combination PPP is implemented.(3)Static simulation dynamic experimental analysis.Based on MATLAB,the algorithm of multi-system combination PPP is implemented,and the data is analyzed and verified.The experimental results show that the accuracy of the random walk model and the constant acceleration model is generally the same in the static environment,but the convergence speed of the CA model is better than the random walk model;under two dynamic modes,The GPS/GLONASS/BDS three-system combination can improve the dynamic PPP positioning accuracy and accelerate the convergence speed.The plane average positioning accuracy of the CA model is 3.4cm,and the average convergence time is 15.88min.The plane average positioning accuracy of the random walk model is 3.5cm,and the average convergence time is 16.63 min.(4)Dynamic on-board experimental analysis.Through multiple systems combination Dynamic PPP experiment,the experimental results show that multi-system combination can accelerate convergence and improve positioning accuracy.The GPS/GLONASS/BDS three-system combination can achieve a positioning accuracy of 4.4 cm in the N direction and 6.3 cm in the E direction in 48.7 min.Figure[24]Table[9]Reference[101]...