RU2010151023A - METHOD FOR SUPPORTING THE AIR GOAL OF THE HELICOPTER CLASS - Google Patents

Abstract

 A method of tracking an air target of the “helicopter” class, which consists in calculating the procedure for the optimal multidimensional linear discrete Kalman filtering described by expressions! ! ! ! ! ! ! where k = 0, 1, ... is the filter cycle number; ! P- (k + 1) and P (k + 1) are the covariance matrices of extrapolation and filtering errors, respectively; ! Ф (k) - transition state matrix; ! Q (k + 1) and R (k + 1) are the covariance matrices of excitation and observation noises, respectively; ! K (k + 1) is the matrix of weights; ! I is the identity matrix; ! and - the vector of current and extrapolated estimates of the range to the target and the Doppler frequency, due to the speed of approach of the air target with its tracking station; ! H (k) is the observation matrix; ! Y (k) is the observation vector; ! Z (k + 1) - matrix of measurement residuals; ! Ψ (k + 1) is the matrix of a priori filtering errors; ! "-1" - inverse matrix calculation operation; ! "t" is the transpose of the matrix,! characterized in that the optimal multidimensional linear discrete Kalman filtering procedure described by expressions (1) - (6) is carried out in parallel in each optimal filter OFmj of their matrix, where m = 1, 2, 3, 4; ! m = 1 corresponds to the stationary nature of the helicopter flight; ! m = 2 corresponds to a helicopter flight with acceleration; ! m = 3 corresponds to a helicopter flight with braking; ! m = 4 corresponds to a helicopter flight in the “hover” mode; ! j = 1, L; L - the number of options for the dynamics of the flight of the helicopter for each of its m-th nature of the flight,! for various a priori data adopted during filtering in each OFmj relative to the mth character of the helicopter flight and the corresponding j-variant of its dynamics for each character of the flight, at

Claims (1)

A method for tracking an aerial target of the “helicopter” class, which consists in calculating the optimal multidimensional linear discrete Kalman filtering procedure described by the expressions

where k = 0, 1, ... is the filter cycle number; P ^- (k + 1) and P (k + 1) are the covariance matrices of extrapolation and filtering errors, respectively; Ф (k) - transition state matrix; Q (k + 1) and R (k + 1) are the covariance matrices of excitation and observation noises, respectively; K (k + 1) is the matrix of weights; I is the identity matrix;

and

- a vector of current and extrapolated estimates of the range to the target and the Doppler frequency due to the speed of approach of the air target with its tracking station; H (k) is the observation matrix; Y (k) is the observation vector; Z (k + 1) - matrix of measurement residuals; Ψ (k + 1) is the matrix of a priori filtering errors; "-1" - inverse matrix calculation operation; "t" is the transpose of the matrix, characterized in that the procedure of optimal multidimensional linear discrete Kalman filtering, described by expressions (1) - (6), is carried out in parallel in each optimal filter OF OF _{mj of} their matrix, where m = 1, 2, 3, 4; m = 1 corresponds to the stationary nature of the helicopter flight; m = 2 corresponds to a helicopter flight with acceleration; m = 3 corresponds to a helicopter flight with braking; m = 4 corresponds to a helicopter flight in the "hover" mode; j = 1, L; L is the number of variants of the dynamics of the flight of the helicopter for each of its m-th nature of the flight for various a priori data adopted during filtering in each RP _mj with respect to the mth character of the helicopter flight and the j-variant of its dynamics corresponding to each character of the flight, while filters are arranged along the rows of the matrix of optimal filters in which dynamic data are accepted as a priori models for various hypotheses regarding the m-th nature of helicopter flight, and in the columns, filters with dynamic models for various hypotheses regarding jx helicopter flight dynamics options with the corresponding its mth character of flight, for each optimal filter of their matrix, the corresponding values of random variables are calculated

according to the expression comparing the obtained values of random variables

with its corresponding boundary values χ ² _{gr m} (m, Р _ош ), identical for all optimal filters located in the mth row of their matrix,

where Р _Ош is the probability of the error that the correct hypothesis regarding the mth character of the helicopter flight will be rejected, the maximum row number of the matrix of optimal filters is determined, where one or more optimal filters are located for which condition (8) is fulfilled, which corresponds to

character of helicopter flight, for those optimal filters OF

for which in

-th row of their matrix, condition (8) is satisfied, the corresponding values of the generalized variances of the real filtering errors are calculated in accordance with the expression

determined by the column number

in line

where is the optimal filter for which the value

minimal, which corresponds to the assessment

options for helicopter flight dynamics for the estimated value

the nature of his flight, based on values

and

evaluation selection

from the output of only one OF

from their matrix at the intersection of the estimated row numbers

and column

.