Research On Low-cost Realtime Multi-system PPP/INS Tightly Integration Navigation Key Technologies Under Harsh Environments

The Global Navigation Satellite System(GNSS)is not only widely applied in military,but also promotes progress in the fields of navigation surveying,geological exploration,disaster monitoring,etc.As the fundamental of autonomous vehicles and the intelligent transportation service,land vehicle navigation is one of the important application fields of GNSS.GNSS precise positioning is a key component in autonomous vehicle safety,which enables autonomous vehicle automatically reaching the destination.However,the accuracy and reliability of GNSS can be degraded greatly in GNSS challenging environments(e.g.urban canyon,tunnels,overpasses,etc.)due to the blockage of satellites and degradation of satellite signal quality.Therefore,to meet the miniaturization,low cost,and high reliability requirements of land vehicle navigation,the integrated system of GNSS and Micro-ElectroMechanical System(MEMS)Inertial Measurement Unit(IMU)is one of the important ways to improve the availability and reliability in harsh environments.Based on the analysis of the development of low-cost inertial navigation,singlefrequency PPP and integrated navigation technology,this thesis focuses on the key technology of low-cost multi-constellation GNSS Precise Point Positioning(PPP)and consumer grade Inertial Navigation System(INS)aided integrated navigation.PPP adopts one GNSS receiver to achieve high accuracy positioning,which becomes increasingly widely applied in different areas.The main research works of this thesis are as follows:To achieve the real-time requirement of land vehicle navigation,PPP technology and International GNSS Service(IGS)Real-time Service(RTS)products are systematically and comprehensively analyzed,and the multi-constellation real-time PPP positioning algorithm is implemented.The low-cost multi-constellation PPP overperforms PPP with one constellation in accuracy,which has been verified by performance evaluation.Moreover,the theoretical differences between the tightly coupled integration and the loosely coupled integration in terms of positioning domain and observation domain are explained.Based on the refined processing model of tightly coupled multi-constellation PPP/INS,loose integration in positioning domain and tight integration in observation domain are proposed.Aiming at solving the problem of abnormal GNSS observations in GNSS challenging environments,GNSS outlier detection with the aid of INS is proposed and implemented to reject the abnormal observations.The position update of INS is used in detection statistics,which can effectively detect abnormal GNSS observations.The adaptive Kalman filter based on observation residuals is applied to deal with the abnormal GNSS measurements that may affect the filter stability and estimation accuracy,resulting in reducing the influence of poor observation quality on the filter.A fast matrix inversion method based on Gaussian elimination is proposed to improve the calculation speed of Kalman filter for real-time applications.To solve the problem of invalid positioning caused by insufficient GNSS satellites and quick drift of low-cost MEMS,a velocity constraint method is proposed and implemented to reduce the accumulation of INS errors.A mitigation method based on velocity constraint is proposed by error analysis of PPP/INS tightly coupled integration in harsh environments.Field tests with u-blox M8 U GNSS and IMU raw observations were conducted to verify the performance of PPP/INS tight integration model.The results show the effectiveness of velocity constraint method to limit the quick accumulation of low-cost MEMS errors.Moreover,the performance of low-cost integrated navigation system with insufficient satellites can be improved.A low-cost multi-constellation PPP/INS system for real-time land vehicle navigation is designed and implemented in terms of software and hardware to meet the real-time system requirements.The GNSS receiver and MEMS are synchronized using the second-order holder.The adaptability analysis of different IGS RTS products is carried out in field tests,and suitable products are selected and applied to reduce the positioning error.The proposed system is tested and verified by multiple filed tests.The results show that the continuity and reliability of positioning can be improved with the proposed system in GNSS challenging environments.