DESIGN OF CONTROL SYSTEMS WITH UNCERTAIN PARAMETERS

A method has been presented for designing control systems with uncertain parameters and noisy state measurements. As in optimally sensitive design, the control consists of a nominal plus correction terms which take into account small changes in system parameters and initial conditions. However, in this paper, the correction terms involve estimates of the parameters and initial conditions obtained by means of the Kalman-Bucy theory. It has been shown that the new control tedsto the optimally sensitive control when measurement noise tends to zero. Furthermore, it has been shown that the feedback correction term tends to zero when the measurement noise becomes arbitrarily large, thus making the control tend to open loop control.