GNSS Positioning Based On Nonlinear Filtering And Multi Satellites Information Fusion

Global Navigation Satellite System(GNSS)is applied widely in more and more field,such as vehicle navigation,cellphone navigation,geo-location,and so on.Global Positioning System(GPS)is the most perfect satellite navigation system which can provide continuous three-dimensional position and velocity information in real-time.In recent years,with the advancement of full recovery plan of constellation ability,and the developing and launching of new GLONASS-K/KM/NG satellite,Russian satellite navigation system is developing greatly.Beidou satellite navigation system(BDS),as an independent satellite navigation system independently developed by China,is similar to GPS,which has characteristics such as passive positioning,global coverage,all-day,good concealment.Since the completement of the first phase of BDS by the end of 2012,22 satellites has been launched.BDS can realize the all-day satellite navigation services in Asia Pacific,which will realize global navigation in future.EU is developing Galileo satellite navigation system(Galileo).On the one hand,multi-system promptes the common development and international cooperation.On the other hand,it provides users with more available satellites,which improve accuracy and stability of the positioning.According to statistics,the quantity of various satellite navigation systems which contain in the use,under construction and construction in the plan will be close to or more than 140.How to carry out effective and high precise positioning by using these satellites from different navigation systems is a new and challenging research subject.In this thesis,the main work includes the following several parts:1)Research background and developments of multiple satellite navigation system fusion were summarized in this thesis.Details concerning the states of development,system compositions,time systems,references coordinate systems of GPS,BDS,GLONASS and Galileo.The ways to unify their coordinate systems was introduced after analyzing the similarities and differences of the time systems and coordinate systems of the various GNSS systems.2)The basic positioning principle,important concept in satellite navigation and the most widely used iterative least squares(ILS)estimation were introduced.And then the new ILS method for multi GNSS systems combined positioning based on the unified coordinate system was brought up.3)Analyzing error factors that affect the GNSS positioning,and building state model and measuring model for nonlinear filter.4)Applied the traditional method ILS and the nonlinear filter method to the structured multi GNSS positioning model for high accuracy positioning estimation.The nonlinear filter method is more reliable than least squares iterative estimation.5)The new filter method was proposed by analyzing the advantages and disadvantages of the nonlinear filter method.Using the proposed positioning estimation method,the positioning results were estimated and compared.6)Simulating the insufficient number of satellites in a single system environment,such as urban canyons and mountains canyons,because very few satellite number visibly,then the single system may not position,using multiple positioning system based on nonlinear filtering algorithm can still work normally and get a better positioning accuracy,verify the multimode fusion positioning stability and superiority.7)For deficiencies of existing satellite selection algorithms for multi systems,a new satellite selection algorithm was proposed based on optimum constellation configuration and underlying center method,the satellite selection is proposed for the characteristics of multi systems and validated.Compared and analyzed with the other satellte selection methods,it is illustrated that the method based on optimum constellation configuration and underlying center in this thesis can achieve a balance between accuracy of positioning and computation.