UOGOLNIONA ZASADA STOCHASTYCZNEJ ROWNOWAZNOSCI I JEJ ZASTOSOWANIA.

Stochastioc optimal controls of discrete linear systems with quadratic performance indices and disturbances belonging to a certain general class of stochastic processes are considered. A proof is presented showing that the optimal strategy for the case of incomplete information results from that derived for the case of complete information. It is accomplished by replacing the variables with their estimates. The mean value of the performance index determined for the case of incomplete information is compared with that evaluated for the case of complete information. Applications of the generalized principle to problems involving different types of disturbances as well as available inform are discussed. It has been shown that the well-known separation theorem is a special case of the generalized principle. The case of Gaussian correlated disturbance is also considered.