Minimax problem of estimating trajectory parameters in continuous formulation

The problem of the linear estimation of trajectory parameters in a continuous formulation is considered in the case when the measurement errors contain two mutually noncorrelated components; (1) white noise, and (2) a random process with a known variance and an unknown correlation function. A formulation is given to the problem of determining the unbiased linear estimation algorithm from the condition of the minimization of the maximum according to the unknown correlation function of the variance of the derivation of the estimate from the true value of the estimated parameter. The example of the nearly circular motion of a satellite is considered.