Аннотация
A method of solution is proposed for a discrete linear-quadratic optimal control problem which consists of aggregating components of the state variables and of the dual variables. This generates another problem with fewer state variables. A two-stage iterative scheme is given for updating the aggregation parameters and a theorem is stated guaranteeing convergence of the sequence of original state and control variables corresponding to the sequence of aggregated solutions, to the optimal solution of the original problem. Ключевые слова
discrete linear-quadratic optimal control; two-stage iterative scheme; aggregation |