Research And Implementation Of GNSS/SINS Integrated Navigation System

With the rapid development of modern science and technology,navigation technology has become a hot area because of its high precision,high dynamics,and high reliability.Among them,Global Navigation Satellite System(GNSS)is becoming more and more mature after more than 50 years of development.However,single GNSS has some problems,such as the limitation of application environment and poor anti-interference ability.The combination of GNSS and Strapdown Inertial Navigation System(SINS)can effectively improve their respective shortcomings and significantly improve the overall navigation performance of the system.In this paper,according to the complementary characteristics of GNSS/SINS integrated navigation,the key technologies are discussed,and the GNSS/SINS integrated navigation and positioning system is designed and implemented.The main research work and innovative achievements of this paper are as follows:Firstly,on the basis of the research background and status quo of GNSS,SINS and integrated navigation,the basic theory of GNSS/SINS integrated navigation system is introduced.It mainly includes: the multiple spatial coordinate systems used in this paper and their mutual transformation relationship,the basic principle and positioning method of GNSS,the basic principle and updating algorithm of SINS,and the comparative analysis of three different combination methods.Secondly,the error model of SINS is established and simulated,including deriving the error equations of velocity,position and attitude,the components and sources of SINS-IMU errors are analyzed,and the distribution of random errors is obtained by analyzing the measured IMU data with Allan variance method.The static error propagation of SINS is simulated,and the results show that gyro bias and random error are the main error parameters that affect the navigation accuracy of SINS.Then,on the basis of different error parameters of SINS,the GNSS/SINS loose combination system equation is established,and the classical Kalman filter algorithm is used to verify its simulation.Four common nonlinear filtering algorithms are analyzed and compared,which provide different algorithm choices for different application environments of integrated navigation.Aiming at the problem that GNSS is often out of lock,which leads to filter divergence in integrated navigation system,an auxiliary model based on LMBP neural network is proposed.The simulation results show that the model can restrain filter divergence in GNSS out of lock and improve the positioning accuracy of integrated navigation system.Finally,based on the theoretical knowledge of integrated navigation,the hardware board of GNSS/SINS loose integrated navigation system was designed.The control core of the hardware is Arm Cortex-M4,the receiver of GNSS is UBLOX?M8T,and the inertial device of SINS is ADIS16405.In order to better real-time monitor whether the data changes of the integrated navigation system is consistent with the actual situation,the GNSS/SINS Integrated Navigation visualization monitoring software is also developed.On this basis,the actual vehicle experiments were carried out to verify the effectiveness and accuracy of the loosely integrated navigation system designed in this paper.The research results of this paper have strong practical engineering application value.