Control research of network servo system based on Smith predictor

Servo motor can be closed-loop controlled through the network, it brings benefits to some long distance control systems. But the network inevitably introduces time delay to the system. To solve this problem, network servo system based on Smith predictor was designed. Control strategy combines the adaptive least mean square (LMS) delay prediction algorithm and Smith compensator can compensate a certain degree of the network delay. Moreover, the combined algorithm is debugged in a real-world system with an interface that developed by the authors. The experiment results show that: in contrast with traditional PID algorithm, PID network servo system with Smith compensator has little effect by network delay. So the method we proposed shows much more robustness and self-adaptability than the traditional PID algorithm.