Robust controller synthesis of discrete-time intelligent systems using a standard neural network model

The design of robust stabilizing controller for discrete-time standard neural network model (SNMM) is investigated. The sufficient condition is given for the existence of state feedback robust stabilizing control laws by uses of Laypunov stability theory and S-procedure technique. A parameterized representation of the control laws is given in terms of linear matrix inequality. Most neural-network-based (or T-S fuzzy) discrete-time intelligent system can be transformed into the discrete-time SNNM for controller synthesis in a unified way. The numerical examples and simulation results show that the presented method is effective and provide a new approach to the synthesis of the controllers for other nonlinear system.