Strategy Research For Improved Precision Based On Distributed Inertial Navigation System

Inertial navigation system plays an extremely important role in modern militarized war with its unique advantages of autonomy and concealment.With the wide application of high and new technologies in the military field and the constant reform of military thoughts,the modern war pattern has undergone profound changes,and the accuracy of local navigation information is required to be more strict.Therefore,a distributed system of meshing low-precision inertial measurement unit with high-precision inertial navigation(or integrated navigation system)is proposed.By sharing inertial network node information,accurate local state estimation and global performance optimization technology,local reference accuracy of airframe weapon hanger,radar and other task nodes can be improved,and global deformation information of airframe can be provided to support the health diagnosis of airframe structure.This paper takes the distributed inertial navigation system as the research object,analyzes its principle and model,and focuses on improving the navigation accuracy of the system.The main research contents include:1.Inertial navigation technology and related theories are studied.Firstly,the relevant terms of inertial navigation,definition and transformation between coordinate systems are introduced.Then the working principle of the strapdown inertial navigation system is introduced in detail with the geographic coordinate system of "east-north-sky"(ENU)as the navigation coordinate system.Finally,the discretization algorithm and error equation of the system are derived.2.After the working principle of inertial navigation system is described in detail,the precision improvement strategy of node transfer alignment is discussed based on kalman filtering principle.Firstly,the basic kalman filtering theory is introduced,then the principle of transfer alignment and the matching mode are explained,and the kalman filter model is established based on the matching method of velocity plus attitude.However,since the convergence of kalman filter needs a certain time,the filtering accuracy is low before the convergence,but in fact,it needs to have a high precision estimation result in the whole flight process.Finally,a reverse kalman filter model is established to optimize the estimation results of node transfer alignment by combining forward and backward filtering,so that the navigation accuracy can also have better performance in the initial stage of navigation solution and the transfer alignment performance can be improved in the whole time.3.From the point of view of distributed structure,the navigation precision of multiple nodes can be improved.Firstly,This paper first introduces the structure and characteristics of the distributed model,then describes the various navigation information association models of multiple nodes in the distributed inertial network,and then determines the initial relative postures between two nodes by using the TRIAD(ThRee-axIs Attitude Determination)to complete the preliminary work of subsequent information fusion.The use of relative attitude to complete each node associated the transfer of information,using information fusion algorithm improve the network structure of inertial navigation accuracy is,the sub-nodes of the navigation precision of estimation results than single node the state estimation precision of transfer alignment get improved,this paper provides a feasible method to improve the navigation accuracy of the distributed system.