Naidenov V.G.

In practice of carrying out of test experiments there is a problem of joint mathematical processing multichannel trajectory the measurements received from territorially carried measuring instruments which can have mutual time are nonviscous in a binding of these measurements to a uniform time scale. The times nonviscous between the moments of time of a binding to a uniform time scale trajectory a measuring complex of the identified measurements leads at their joint statistical processing to additional errors in an estimation of a vector of parametres of movement of tested flying objects. Carried out before research have shown, that decrease in influence of time errors between the moments of time of a binding to a uniform time scale of the identified measurements on accuracy of an estimation of a vector of parametres of movement of flying objects can be carried out with use of specially synthesised modified filter Kalmana using linear model of observable process in which the additional error of measurements caused only in the speed of change measured means trajectory of measurements of nonsynchronous parametres is considered. However in practice for dynamical and highly maneuverable flying objects it is necessary to consider as well accelerations of change measured by means trajectory measurements of parametres. Absence of such possibility leads, in some cases, to additional errors in an estimation of a vector of parametres of movement of flying objects by results of high-precision measurements. Thus the model of vector casual process trajectory measurements becomes obviously nonlinear in relation to a vector of time errors in a binding of measurements to a uniform time scale. The method and the programmno-realised algorithm of a suboptimum nonlinear filtration of multidimensional stochastic processes of measurements of trajectories of highly maneuverable flying objects in the presence of time errors in a binding a component of such processes to a uniform time scale have been with that end in view developed. Such algorithm has been realised on the basis of classical parities the Kalman,s filtrations, the developed procedure linearizathon vector dependences and by the account in parametres of the synthesised filter of the additional aprioristic information on stochastic characteristics of time errors in a binding primary trajectory measurements to a uniform time scale. Research of the developed method of a suboptimum nonlinear filtration of multidimensional stochastic processes of measurements of trajectories of highly maneuverable flying objects in the presence of time errors in a binding a component of such processes to a uniform time scale has been spent on an example of the analysis of results of two measuring experiments. Results of approbation confirm high efficiency of the offered method of a suboptimum nonlinear filtration of multidimensional stochastic process of measurements of trajectories of highly maneuverable flying objects in case of presence of the additional aprioristic information on stochastic characteristics of errors in a binding a component of such process to a uniform time scale that gives the basis to recommend it for use at processing of results trajectory measurements.