Singular Value Decomposition Approach to in-flight Alignment of Strapdown Inertial Navigation Systems

摘要

Observability and degree of observability analysis of a dynamic system is an essential procedure to determine the efficiency of a Kaman filter designed to estimate the state of that system. A singular value decomposition method of observability matrix of a dynamic system is proposed. By means of this method, advantages are achieved for the degree of observability analysis of strapdown inertial navigation system (SINS) in initial alignment stage, and the choice of optimal carrier vehicle maneuvers trajectory is determined in initial alignment stage for the best efficiency of Kalman filters. The efficiency of this analysis method is demonstrated by the computer simulation results.