On Investigation Of GPS/SINS Tightly Integrated Navigation System

Global Positioning System(GPS) provides high accuracy of positioning output long term, its error does not diverge over time, but the satellite signal is easily attenuated,multipath effcets has significant influence on GPS under complex environment, and thus limits the navigation precision. Strap-down Inertial Navigation System(SINS) have advantages such as being independent of external information and it can work in any environment. But the error of SINS will diverge over time and can not work independently long time. GPS and SINS can complement each other, which make GPS/SINS integrated system has excellent navigation performance, and the performance of tightly integrated navigation is better than the two separate systems. This paper study the GPS/SINS tightly integrated navigation system, analysis the existing problems, put forward the solution methods to improve navigation precision.Firstly, this paper introduce basic principle and composition of GPS and SINS respectively, then study the GPS/SINS integrated system, mainly the tightly integrated navigation system, and establish the mathematical model of the system. Because of the nonlinear mathematical model of tightly coupled system, the extended Kalman filter is suggested in this paper. On this basis, integrated Kalman filter is designed.Secondly, this paper is concerned with the GPS satellite selection algorithms, and pays particular attention to the satellites selection efficiency of GPS system. On basis of tetrahedron volum method, a new selection method is proposed in this paper, the new method improve the satellites selection time and efficiency. Through simulation to verify when the number of satellite less than 4, tightly integrated navigation system can work normly.Finally, this paper analyzes the influence of time synchronization errors on GPS/SINS tightly integrated navigation system, studies the case when the time synchronization errors are not included in the integration filter, which lead to an increased error covariance and the inaccuracy on SINS basis estimation, put forward expanding time synchronization error to the state vector, and analysis the observability of time synchronization error. Through simulation the feasibility of the method is proved.