Model Predictive Control for Networked Control Systems With Random Data Dropouts
In this paper, the model predictive control for the networked control systems is proposed. The problem of data dropouts in the sensor controller link when the controller does not receive new feedback data is treated. As in predictive control design the sequence of future control actions up to a given horizon is calculated at each sampling time, the natural idea is to use not only the first control action, but also other terms of the control sequence, in case the sensor data at next sampling times are not available. From the implementation point of view there are two or more control laws (depending on the number of lost output samples) which are switched arbitrarily fast.