New Smith Predictor and FRBF Neural Network Control for Networked Control Systems

This A novel approach to random, time-variant network delay in the networked control systems (NCSs) and a controlled plant is proposed in this paper. The approach is a new Smith predictor combined with fuzzy radial basis function neural network (FRBFNN) control. It can identify the controlled plant and adaptively adjusts weights of the controller. Because new Smith predictor hides predictor model of the network delay into real network data transmission process, further the network delay no longer needs to be measured, identified or estimated on-line. Simultaneously, this new Smith predictor doesnpsilat include the prediction model of the controlled plant, thus it doesnpsilat need to know the exact mathematical model of the controlled plant beforehand. Thus, it is applicable to some occasions that network delays are the random, time-variant or uncertain, possibly large compared to one, even tens sampling periods, and there are data dropouts in the closed loop. Based on CSMA/AMP (CAN bus), the results of the simulation show the validity of the control scheme.