Aerospace Instrument-Making Annotation << Back Integration of inertial-satellite navigation systems on the basis of the stochastic filter, invariant to object model Sokolov S.V. Yugov Yu.M. We consider the basic possibility of solving the problem of integration of inertial- satellite navigation systems (NS) based on the theory of stochastic filtering without using the model information of the object, the nature of his motion, etc. Equations of vector navigation parameters of the moving object on the basis of a measuring complex strapdown system, that are invariant to the nature of the object and type of disturbances acting on it, ensure the possibility of a posteriori optimal estimation of motion parameters with the loss of satellite communications using as an observer state vector magnetometer are considered at the work. For the regime of satellite measurements equations of the linear parameters of the object motion that are invariant to the form of the object model, etc., are derived from the vector Doppler measurements and allow us to solve the problem of optimal estimation of a posteriori coordinates of an object when used as an observer state vector code measurements. In this case, the estimation of the angular parameters is similar to an autonomous regime estimates from the measuring complex strapdown system. Despite the fact that these equations of the tight-coupled NS allow optimal to estimate navigation vector for reduce the computational cost in article designed suboptimal Kalman estimation algorithms, the invariant to model of the object and provides stable and highly accurate estimation of the angular parameters and coordinates of the object, as with a satellite measurement, and at their failure. Key words: navigation algorithms, the tight coupling inertial-satellite systems, stochastic filtration. Contacts: E-mail: jura-jugv@rambler.ru Pp. 09-17.