Guaranteed optimization of uncertain systems through the use of prediction

A new type of special optimal state feedback is introduced for optimizing d ynamical systems under conditions of constantly acting disturbances. Using a priori and current information, possible disturbances are represented as the sum of sufficiently regular and irregular components. An algorithm of o peration of a controller is described; this controller processes the trajec tory executed and calculates the values of an optimal feedback control for each particular control process in a real-time mode. The results are illust rated by an example.