Minimum entropy control of closed-loop tracking errors for dynamic stochastic systems

In this paper, a minimum entropy control approach for closed-loop tracking error is studied for linear dynamic control systems subjected to stochastic disturbances. The basic idea is to construct a feedback control scheme with the guaranteed closed loop stability using the Youla parameterization that provides a set of free parameters embedded in the controller. The entropy of the closed loop tracking error is then minimized by selecting these free parameters in the Youla parameterization. In this context, the Renyi’s entropy with the Parzen window estimation is used to measure the information contained in the tracking error. Such an entropy measure characterizes the uncertainty of the closed loop control system. By minimizing such entropy measures, the selection of these free parameters has been formulated. An illustrative example is included to demonstrate the use of the control algorithm, and satisfactory results have been obtained.