ABSTRAKT

Nonconvex set of values of parameters consistent with the current measurement of the input and output variables is represented as the union of a finite number of convex subsets. The solution of the estimation problem is reduced to choosing one of the subsets and solving the obtained system of convex inequalities. This problem is solved using the previously proposed ellipsoid method modification which works in case of a finite number of incompatible inequalities. Computational cost of the proposed algorithm is compared with that of similar estimation problem without noise. The properties of the algorithm are illustrated by numerical examples.