On the need for and use of a measure of state estimation errors in the design of quadratic-optimal control gains

In this correspondence it is shown quantitatively that state estimation error acts like an additional disturbance on the stochastic linear regulator. The closed-loop system´s covariance matrix is shown to be an explicit function of the optimal filter´s state estimation error covariance matrix. Ways in which this information can be used in design of the feedback gain matrix of the deterministic linear regulator are discussed.