Design of robust networked predictive control systems

Abstract

Random communication time delay highly degrades the control performance of networked control systems. The paper addresses the problem by presenting a novel control technique using a modified model predictive control (MPC) and a modified Smith predictor. By using MPC, future control sequences are generated to compensate for the forward communication time delay when control signals are delayed or missed. Using the Smith predictor, the backward time delay can also be compensated for. The design of a filter is used to improve the robustness of the networked control system. To illustrate the improvement in control performance using the proposed method, simulation results are presented to show that the method is superior to the standard GPC and networked predictive control.