Research On Ambiguity Resolution Of INS-aided High-precision GNSS In Urban Environment

With the rise of autonomous driving technology and the blowout development of unmanned drones,unmanned vehicles as well as mobile robots,the industry applications represented by mobile surveys and the mass market applications represented by location services have a huge demand for precision positioning.The Global Navigation Satellite Systems(GNSS)and Inertial Navigation System(INS)integrated navigation system can incorporate the benefits of each other and overcome their deficiencies.Through complementary advantages,the adverse effects of the complex urban environment can be effectively overcome,and the continuous positioning capability of the integrated navigation system can be improved.Facing the increasingly high-precision navigating and positioning requirements in a complex urban environment,this research takes GNSS as the main research subject and conducts corresponding research on dynamic high-precision GNSS positioning models.To improve the availability and reliability of GNSS positioning solutions in complex urban environments,a high-precision GNSS/INS integrated navigation system was constructed by introducing inertial sensors and combining them with improved mathematical models as well as optimized fusion methods.Based on the ambiguity resolution(AR)characteristics of different high-precision GNSS positioning models,the corresponding improved and enhanced algorithms were proposed to improve GNSS positioning accuracy and enhance GNSS continuity and reliability.The main work and contributions of this dissertation are as follows:(1)The user-side PPP-AR mathematical models based on the integer-recovery clock(IRC),fractional cycle bias(FCB)products,and phase clock(PC)products were derived,and a unified multi-constellation PPP-AR data processing based on different product forms was established.In addition,BDS and Galileo observations were added to establish multi-GNSS PPP-AR to accelerate the convergence of GPS-only PPP-AR.The positioning performance of multi-constellation PPP-AR with different analysis center products was evaluated and the feasibility of PPP-AR in the urban vehicle environment was verified.(2)The models of PPK/INS tightly coupled integration,IF-PPP/INS tightly coupled integration,and UC-PPP/INS tightly coupled integration was deduced in detail.From the perspective of INS-aided ambiguity resolution,the equivalence between the direct method ambiguity resolution and the constrained method ambiguity resolution in PPK/INS tightly coupled integration was verified.(3)Aiming at the multipath error of pseudorange(PR)observations and the frequent occurrence of cycle slips of carrier phase observations in complex urban environments,a robust algorithm based on PR innovations was used to weaken the influence of PR gross error on float ambiguity estimation,the cycle slip detection algorithm based on phase innovations was realized to accurate detection and repair of small cycle slips.An INS-aided PPK quality control strategy was constructed to eliminate or weaken the influence of abnormal observations on ambiguity resolution.The practical performance of PPK/INS tightly coupled integration in a complex urban environment was improved.(4)To enhance the ambiguity fixing performance in complex urban environments,a partial ambiguity resolution was proposed.Based on the decorrelated ambiguity vector,the PAR method can automatically try to select an ambiguity subset that can satisfy both the requirement of the model-driven bootstrapping success rate and datadriven Fixed Failure-rate Ratio Test.Combined with the PPK quality control algorithm aided by inertial information,a PPK/INS tightly coupled integration with robust and partial ambiguity resolution considering small cycle slip was constructed to achieve more reliably ambiguity resolution.(5)The theoretical consistency between the ambiguity-fixed UC-PPP and the ambiguity-fixed IF-PPP was demonstrated,and the ambiguity-fixed PPP/INS tightly coupled integration unified data processing procedure based on UC-PPP and IF-PPP model was deduced.According to the characteristics of dynamic GNSS data,a progressive quality control strategy was proposed to improve the ambiguity resolution performance of the PPP/INS tightly coupled integration system.Then,a multiconstellation ambiguity-fixed UC-PPP/INS tightly coupled integration model was constructed to enhance the continuity and stability performance of PPP-AR in the urban environment.(6)Open source the multi-constellation GNSS PPP ambiguity resolution program and GNSS/INS tightly coupled integration program as well as the corresponding data set.