IoTにおけるデータ伝送遅延補償制御

抄録

In the network communication there are delay, packet loss, and jitter. Dely is most important for networked control systems, since this factor occurs the most frequently. When there is a delay in the feedback loop, the controlled system tends to be unstable. On the other hand, it is well known that Smith predicter is amazingly effective for a plant with long dead time. Thus, in this paper applicability of Smith predicter to networked control systems is examined. It is verified that this method is highly effective to stability of a networked control system by simulations. However, there is a delay on controlled response due to network time-lag. A real-time control cannot be achieved by Simith predictor. To compensate this delay, linear extrapolation is adopted. As a result of the simulation by using Smith predictive control and linear extrapolation, the proposed method is useful for real-time networked controlled systems or IoT.