Research On RTK Positioning Technology And APP Development Of Smartphone

In recent years,with the wide application of Global Navigation Satellite System(GNSS),demand for the number and accuracy of GNSS equipment has grown,resulting in the development of high-precision positioning of low-cost GNSS equipment.On the other hand,as smart terminals and chip technology have advanced,the satellite navigation and positioning capabilities of smartphones have increasingly gained prominence,and they have become the most widely used type of low-cost GNSS equipment.China is currently aggressively promoting the research and application of GNSS,and the Ministry of Industry and Information Technology of the People’s Republic of China has issued "Several Opinions on the Promotion and Application of Beidou in the Field of Mass Consumption",which explicitly proposes pilot demonstrations of highprecision smartphone positioning.In this regard,this thesis focuses on the application of high-precision positioning on smartphones,and take full advantage of Google’s opportunity to open the raw GNSS measurements of Android devices to systematically conduct research on smartphone highprecision positioning data processing methods.The research primarily focuses on the quality assessment of smartphone GNSS observation,the Real Time Kinematic(RTK)algorithms suitable for smartphone,as well as the implementation of smartphone RTK application(APP).The main contributions include:(1)Comprehensively evaluate the quality of GNSS observations of Xiaomi Mi 8 smartphone and propose a set of data quality control schemes.The double-difference recovery single-difference method is introduced to calculate the observation noise.The GNSS data quality of Xiaomi Mi 8 in open,occluded,electromagnetic interference,and other environments are analyzed through self-programming,using the carrier-to-noise ratio(C/N0),multipath,cycle-slip ratio,data integrity rate,and single-point positioning accuracy as indicators.And data quality control is controlled according to the idea of direct and indirect.The results show that the pseudorange noise of the smartphone is about 5～10 times that of the survey-grade receiver,and the carrier noise is about 2～5 times that of the survey-grade receiver;The multipath,cycle slip,and data loss are all problematic in various scenarios,particularly for the L5 frequency,and the real availability of dual-frequency data is poor.Using the LLI(Loss of Lock Indicator)and Doppler output from smartphones and the pseudorange-phase combination to jointly detect cycle slips,which is used to directly control data quality,can effectively prevent the occurrence of abnormal positioning solutions.(2)The optimization RTK positioning algorithm suitable for smartphone is systematically studied,and the RTK positioning performance is improved from three aspects: the stochastic model,parameter estimation and ambiguity resolution.Taking C/N0 as the indirect control method of data quality,the traditional SIGMA-ε stochastic model is enhanced,and a SIGMA-Δ model considering the difference of dualfrequency C/N0 as well as a method for adjusting the weights based on the moving window are proposed.The results show that the three improved stochastic models can effectively improve the positioning accuracy compared to the traditional SIGMA-ε model.A first-order autoregressive colored noise model is established,and a method of combining functional model fitting and robust filtering is proposed to reduce the impact of colored noise in smartphone GNSS observations,based on an in-depth analysis of the existence,types,and main sources of colored noise in smartphone GNSS observations.The results show that applying a filtering algorithm that takes into account colored noise improves accuracy by roughly 30%.A partial ambiguity resolution method is introduced to improve the fixed rate of smartphone RTK,based on the integer characteristics of ambiguity in smartphones,and the results show that the fixing rate can reach more than 90% under static conditions,the fixed solution accuracy is 1-2 cm,and it can accurately identify the centimeterlevel deformation trend.(3)Developed a smartphone RTK positioning APP that was utilized for high-precision navigation,low-precision surveying and mapping.The real-time capture of smartphone high-precision RTK location information is realized,and the visualization function is interactive and friendly,based on an optimized smartphone RTK positioning algorithm using the Java language and JNI(Java Native Interface)technology.The application testing results show that the APP can achieve plane decimeter-level precision in pedestrian and vehicle scenarios,which is better than the smartphone’s own chip solution;When smartphone RTK is utilized for geographic information collecting,the plane accuracy is around 1 m,and in most cases the error is submeter,which is basically meet the needs of low-precision surveying and mapping applications.