Research On GNSS/SINS Navigation And Positioning Method For Vehicle/Ship Mobile Measurement System

With the rapid development of surveying and mapping,new equipment for surveying and mapping has been emerging.Mobile measurement technology has become one of the froniter technologies in the surveying and mapping.In the mobile measurement system,the dynamic positioning is at the core position,which provides the position reference for the data process of other sensors.The global satellite navigation and positioning system can provide high-precision position information,and the error is independent of observation time,but it is affected by the external environment.Inertial navigation system is independent of the environment,can output navigation information alone,but its navigation error will accumulate with the observation time quickly.GNSS/INS integrated navigation can maximize the advantages of both and provide high precision and stability navigation information for mobile measurement.This paper mainly studies the following aspects:(1)Analyze the principle of satellite navigation,deduce the double-difference observation equation of three systems(GPS,GLONASS,BDS).Then give four kinds of common linear combinations.Finally,t analyze the least squares solution and different stochastic models,and compare the accuracy of different stochastic models.(2)Research and study on the single epoch localization algorithm of medium and short baseline,and give a complete single epoch ambiguity search strategy.The main idea of this method is to fix the wide lane ambiguity and then constrain the fixed basic ambiguity.The SNR can reflect the quality of the observation.Therefore,a group ambiguity search algorithm based on SNR selection is proposed.Through the experiment,the method can improve the efficiency of ambiguity search and satisfy the requirement of dynamic positioning.(3)The basic principles of Strapdown inertial navigation are studied,the common coordinate system is explained and the transformation relationship between them is deduced.Then the navigation equations and algorithms in the local navigation coordinate system are deduced,and three kinds of common inertial navigation initial alignment methods are analyzed,including the direct alignment method applicable to the actual engineering.(4)Taking the GNSS/SINS loose combination model as the research object,and the state equation and observation equation of loosely combined model are given.The basic principles of Kalman filtering and robust adaptive filtering are discussed.Aiming at the GNSS/INS loose combination model,it proposes to use prediction residuals to construct adaptive factors and robust factors,so as to adjust the weight relationship between dynamic models and observations,the influence of the model error and the observation error on the filtering accuracy is reduced.Through experimental data analysis,compared with standard Kalman filtering,the improved filtering method can improve the precision and enhance the stability.