Neural controllers for networked control systems based on minimum tracking error entropy

摘要

In this paper, a novel proportional–integral–derivative (PID) feedback control method for networked control systems (NCSs) subject to random delays is presented via minimizing tracking error entropy, which is estimated by Parzen windows and quadratic Gaussian kernels. The PID controller is implemented by backpropagation (BP) neural networks. Specifically, the performance index implies the idea of the minimum entropy control of the closed-loop tracking error. The convergence in the mean square sense has been analysed for closed-loop NCSs. Simulation results are provided to show the effectiveness of the proposed approach.