Study On Initial Alignment Of Strapdown Inertial System

Initial alignment is one of the key technologies which affect the effectiveness of the inertial navigation system. The accuracy and rapidity of initial alignment are greatly concerned with the property of the inertial system. In this thesis, the main work includes the following aspects:1. The error equations of SIS are studied. At first, some kinds of coordinate in the inertial navigation system are defined. Then, the attitude update algorithms of SIS are studied, and the basic dynamics equations based on geography coordinate are introduced. At last, the error equations of SIS are derived.2. Kalman filter and its application in initial alignment are studied. At first, the application background of Kalman filter theory is introduced. Then, discrete Kalman filter equations are derived, and state equations of continuous system are discreted. 2-channels-10-states and 2-channels-12-states error models are presented and are used for the initial alignment of marine SIS on moving base, and measurement models are given using velocity matching and position matching.3. The observability and degree of observability of initial alignment of SIS on moving base are all-around analyzed and studied. The influence of various maneuver of base on the observability of SIS alignment procedure is qualitatively analyzed by means of the observability analysis theory and method of piece-wise constant systems (PWCS). The degree of observability of every state during initial alignment of SIS on various moving bases is quantitatively analyzed using method of singular value decomposition (SVD). At last, many valuable results are obtained, it provides a theoretical basis for devising the best maneuver during initial alignment.4. Simulations of the initial alignment of SIS on various moving bases are performed. The validity of the SVD method is proven by covariance simulation. To obtain the best estimate effect, the combined maneuver is designed. The simulation results show that the error of inertial sensors can be well estimated and the precision of alignment is improved.