作者: P. J. BLANVILLAIN , T. L. JOHNSON
DOI: 10.1080/00207177808922454
关键词: Linear system 、 Quadratic equation 、 Dual (category theory) 、 Mathematics 、 Linear-quadratic-Gaussian control 、 Linear-quadratic regulator 、 Mathematical optimization 、 Optimal control 、 Control theory 、 Control theory 、 Optimal projection equations
摘要: Abstract The steady-state optimal control of a linear time-invariant stochastic system by means minimal-order dual-observer-based compensator is considered in this paper. structure the fixed while associated gains are to be chosen so as minimize quadratic penalty on plant state. Necessary and sufficient conditions for optimality given, an explicit solution displayed. Salient features pertaining : decoupling property, projection innovation property. Finally, it shown that design corresponds singular LQG problem, which precisely dual another problem: namely Newmann's problem. A complete picture then given showing clearly correspondence between two designs.