Research On GFINS/GPS Integrated Navigation System

Today, some modern tactical weapons such as missiles have the characteristics oflarge-overload and high-mobility. The inertial device contains gyroscopes and accelerometersin the common strapdown inertial navigation system, but the gyroscopes has the poor abilitiesto shocks which dicde that the SINS can’t work validly in the high acceleration and highangular velocity conditions. But the Gyroscope Free Strapdown Inertial Navigation Systemcan suit for the high dynamic circumstances because of it is only composed of accelerometers.Especially with the prosperous development of MEMS technology in the last few years, theGFSINS’s advantages such as long-time running, high reliability, low power consumption,low cost, become more and more obvious.Therefore, in recent years many scholars at homeand abroad are doing deep exploration and research on GFSINS, in order to try to put it toengineering and practical application.Compared with the traditional strapdown inertial navigation system, GFSINS navigatesless precisely and accumulates errors faster. Limited by the GFSINS’s navigation theory andthe manufacturing process of the accelerometer, it is very difficult to continue to improve thenavigation accuracy of GFSINS. Therefore, while continuing to develop high precisioninertial navigation systems, integrated navigation system become a focus of the developmentof navigation systems, in which inertial/satellite navigation system has the most gloriousapplication prospect. This article mainly study on GFSINS/GPS integrated navigation system.Due to the integrated navigation system is an extremely complex system, papers focuses on asfollows:First, the feasibility of the Gyroscope free strapdown inertial navigation system is provedfrom the perspective of mathematics. Based on the analysis of the GFSINS’s feasibility, a newconfiguration of nine accelerometers is proposed. The plan of the configuration is designedsimply and the accelerometers are easy to be installed. The redundant information which canbe get from the configuration such as angular acceleration and angular velocity product arebeneficial to the calculation of angular velocity. After get the angular velocity, the process ofthe gyroscope free strapdown inertial navigation is analized and simulated. The simulationresults show that: the single gyroscope free navigation system has the low precision, fast erroraccumulation over time,and can’t satisfy the need of navigation.Second, An extended H_∞filtering algorithm is detailed studied to use to calculate theangular velocity. The simulation results show that: the extended Kalman filtering can be effectively used to increase the precision of the angular velocity calculation only in whitenoise; but the robust filtering also keeps strong robustness under the colored noiseinterference.Third, a new loose combination mode of position, velocity and attitude is introduced intoGFSINS/GPS integrated navigation system. In the new loose combination mode，GPS’soutput of attitude information can effectively correct the GFSINS’s output of the attitudeinformation. After establishing the error model of the integrated navigation system, Kalmanfilter method and H_∞filtering method are applied to navigation calculation under thedifferent noise error model of accelerometers. Then, software simulations are done to contrastthe navigation accuracy under different conditions. At last a H_∞filtering algorithm based ongradually disappear memory is proposed to aim at the problem of model drift of GFSINS/GPSintegrated system, and a H₂/H_∞baced on LMI filtering algorithm is proposed to overcomethe hard of selecting filtering parameter in the common H_∞filtering algorithm; at last, thealgorithms are simulated to verify effectively.