Optimal compensators with pole constraints

In an observer-based control system, the design freedom which remains after pole assignment is used to minimize-in an average sense-a given quadratic performance index on the plant state and control. The class of controllers and observers satisfying pole constraints is characterized through a certain parameter vector. An algorithm is then developed to compute the optimal controller-observer set. An example is worked out to illustrate that the Optimal controller based on estimated state is not, in general, optimal when the true state is used.