method of estimation target-motion parameters
Yu.G. Bulychev, V.A. Golovskoy
For the angular-measuring systems the problem of estimating the target-motion is traditionally solved most intricately inasmuch its substantial nonlinearity supposing the introduction of the appropriate linearization procedures, building of the iteration estimation algorithm and choosing for it a sufficiently good initial approximation.
In such a classical statement the given problem has received a strict solution within the framework of the know statistical estimation method ( for example, the maximum of the aposteriori probability density, maximum of the likelihood function the least square etc.) using one or another suggestions on availability of the appropriate statistical data. However, the analyses of the real conditions of the functioning of many angural-measuring systems shows that the conditions of their operation are frequently characterized by the substantial structural-parametric uncertainly due to the lack of the authentic knowledge on the distribution laws of the measurement errors and their numerical parameters, occurring malfunctions and failures of the apparatus as well as availability of the interferences of the natural and artificial origin.
In the given article in the regularized statement we have developed an estimation method for the target-motion parameters on the basis of the angular-measuring system under the conditions of the uncertainly specified. The method is based on receiving a set of single estimates and appropriate resulting estimate at the expense of using the families of the independent errors, restrictions on the errors norms and variation series for the upper numerical estimates for the target-motion parameters