Nonlinear network-induced time delay systems with online dynamic Bayesian learning

This paper presents a new control approach for nonlinear network-induced time delay systems using online reset control, neural networks, and dynamic Bayesian networks. We construct a state-feedback based nominal control to develop a linearized system model. The reset control and neural network are employed to compensate for system error due to time delay effect. Finally, we achieve probabilistic modeling for the networked control system (NCS) using a dynamic Bayesian network (DBN) to estimate future system outputs and devise a predictive control system. Our control methodology is evaluated through numerical simulation by investigating its control performance and comparison to a conventional state feedback control approach.