Study On Matching Algorithm And Selection Of Suitable Matching Area For Underwater Gravity-aided Inertial Navigation System

The particularity of underwater environment made some techniques such as satellite navigation system,celestial navigation system unavailable for underwater vehicle's navigation.Thus,the Inertial Navigation System(INS)became the most important equipment for underwater vehicle's navigation.But the pure INS has one insuperable shortcomings,because the state information of vehicles is calculated out by utilizing the integral operation mode,which lead to the error accumulation in INS,and ultimately affect the accuracy and reliability,and even the divergence of the results of navigation.This problem can be solved by improving the performance of the gyro and accelerometer in the inertial navigation system,but it's difficult and expensive.Geophysical field aided navigation and positioning method,which include terrain aided navigation,geomagnetic field aided navigation and gravity field aided navigation,obvious has many advantages compared with other navigation method for underwater vehicle's navigation.Taking into account the factor that the former two method has the shortcomings of concealment or the electromagnetic interference from the carrier itself,this paper studies the theory and method of Gravity-aided Inertial Navigation System(GAINS).The specific work are as follows:Firstly,the improved SITAN algorithm is proposed to solve the problem of low precision and poor robustness of the existing SITAN algorithm,that is,the matching algorithm called the correlation SITAN + weight reduction iterative technique(CSITAN+WRIT).The robustness and accuracy of the original SITAN algorithm are improved while the real-time performance of the original algorithm is preserved.Secondly,there are a variety of statistical indicators of gravity field information,which reflect the different characteristics of gravity field.How to synthetically utilize those different aspects of gravity field information has become a problem.This paper uses Principal Component Analysis(PCA)to integrate gravity field's local standard deviation,roughness,slope and correlation coefficients into one singular weighing indicator for the matching suitability of gravity field,which can synthesize different information of different gravity field's characteristic with the original information lost within the allowable range.This undoubtedly provides a reference for the selection of suitable matching area.