Research On Information Fusion Method Of GPS/SINS Deep Integrated Navigation System

With the development of society the accuracy and anti-jamming peformence of navigation and positioning are improving.Single navigation system can not meet the demand,thus integrated navigation system appeared.Combined with each other between the navigation systems can complement each other’s advantages.Thus improving various aspects of performance including positioning accuracy in the system.Recently satellite navigation system and inertial navigation system each has advantages and disadvantages and most navigation systems are jointly constituted by both,then the positioning accuracy and navigation performance significantly improved.Recent years deep integrated navigation system has become a hot research.It based on inertial assisted GPS systems vector tracking loop,mainly including concentrated filtering,distributed coherent deep integrated and distributed non-coherent deep integrated.This paper studies theory and key technologies of GPS/SINS non-coherent deep integrated navigation.Including the introduce of the working principle of integrated navigation subsystem,analysis of different integrates method,inertial GPS tracking loop design,data fusion algorithm design,GPS/SINS deep integrated navigation model construction and performance analysis.Specific studies are as follows:Introduces GPS/SIN navigation subsystem works and different integrates method.Mainly Analyszed various structural pattern of integrated navigation system,construction of SINS/GPS deep integrates is based on receiver vector-based tracking loop,with I/Q signal as the navigation filter measurement information,coherent deep integrate use centralized or cascade filter method;non-coherent deep integrate inherited the tradition link of discriminator,after proportion summing calculate the pseudo-range,pseudorange rate,and as a navigation filter input.In this paper,the parameters of the discriminator output pre-filtered with double filter form of carrier tracking filter and pseudo-code tracking filter.It can to some extent reduce the noise introduced by the nonlinear discriminator and adding the carrier frequency acceleration error optimally estimates of carrier frequency,the carrier phase and code phase.Design SINS system for assisted GPS tracking loop,and optimize the design of the model based on the performance of the tracking loop,dynamic information of SINS introduced does not affect the performance of the carrier tracking loop,enhances the dynamic performance of PLL,the dynamic stress borne by the feedforward channel with inertia auxiliary,so you can reduce the impact of carrier dynamics on the value of the loop bandwidth.In this paper,we established phase jitter error model cased by the phase-locked loop bandwidth.Analyze impact on system performance whith inertial aid.The results show:you can choose a smaller bandwid with auxiliary phase-locked loop,when B = 1Hz,tracking loop corresponding sensitivity 24dB-Hz,enhanced ability of weak signal tracking in system.In this paper,we designed a nonlinear VBKF integrated navigation filtering method,solving the uncertainty of noise model.It based on the Bayesian filter and unscented Kalman filtering method,mainly for the pseudo-range,pseudo-range rate and noise variance time-varying.Set changes in the noise variance to a constant value,sine wave,step wave and square wave model.Based on the simulation results we can see that the filtering method can effectively track the variance changes,and navigation system parameter estimation accuracy than UKF almost 1-fold increase.