On the linear optimal digital servo problem

For linear time-invariant digital systems, convergence of the Riccati equation associated with the optimal servo problem is examined. Necessary and sufficient conditions are established for the existence of finite gains when the command input trajectories may fail to be asymptotically stable.