Error Analysis And Simulation For High Precision Strapdown Inertial System

With the rapid development of technology and the social economy, navigation technology is widely used in various fields. Strapdown Inertial Navigation has become an important part of the navigation system for the simple mechanical structure, the high short-term accuracy, the autonomous navigation and other advantages. In order to enhance the performance and the long-term stability in complex environmental, a comprehensive error analysis for SINS is necessary. This paper study in the algorithm error, measurement error, space misalignment, time synchronization and other issues. We analysis feature of different error and establish their mathematical model, and carried simulations for counting the percentage of different error in the system error.With "Digital Platform" instead of mechanical platform, SINS's accuracy of attitude is important to system. However, in the cone motion or large motor sports environments, the "non- commutative error" makes "Digital Platform" shift and cause seriously accuracy loss. "N+p" sub-samples algorithm is expanding from the traditional frequency cone algorithm and compensates cone error by using previous sub-samples and this cycle's sub-samples. However, by simulation, we found that when the number of sub-samples over 4, the performance of "N+p" algorithm has been unable to increase or even decline, which is inconsistent with the residual estimation formula. Aiming at this problem, this paper found the "N+p" algorithm is effective, but the residual estimation formula of "N+p" is wrong. The performance of "N+p" is overestimated. At the same time we note that the derivation process of "N+p" did not consider the effect of the periodic component. The algorithm has room for improvement. So this paper carried out a detailed analysis of the formation and characteristics of cone error, derivate "N+p" algorithm again and give a residual estimation formula of "N+p".Furthermore, the paper considered on the periodic components of cone error, and proposed a new optimization cone algorithm base on the "N+p" algorithm which use periodic components of cone error.SINS system error sources can be divided to three categories:measurement error, space misalignment and time synchronization. In the low-precision engineering applications, the loss of precision due to measurement error is the largest, but in high-precision applications, Space misalignment and time synchronization can not be ignored. They cause serious error in complex environmental. This paper carried on a detailed analysis and simulation about non-orthogonal error, lever-arm, "size effect" and time synchronization error, and carried simlutions for counting the percentage of different error in the system error.Traditional SINS error model based navigation equations, is an approximate first-order linear model which ignores error coupling terms. In the complex environment of high frequency vibration, large motor, instantaneous shock, this first-order linear model can't accurately describe the characteristics of system error. This paper started from the differential equation of navigation, established a nonlinear error model of SINS, and using UKF filtering algorithm, given a error analysis method for combined system system which has second order accuracy.In engineering applications and theoretical study, digital simulation is an essential part. This article describes the WinForms?GDI+?3D visualization for software development, using track inversion and SINS algorithm builded simulation platform framework for SINS, which provided a solid foundation for subsequent work.