Mixture Basis Function Approximation and Neural Network Embedding Control for Nonlinear Uncertain Systems with Disturbances

A neural network embedding learning control scheme is proposed in this paper, which addresses the performance optimization problem of a class nonlinear system with unknown dynamics and disturbance by combining novel function approximator an improved observer (DOB). We investigated mixture basic (MBF) to approximate system, allows approximation global scope, replacing traditional radial basis (RBF) networks technique that only works locally could be invalid beyond some scope. The classical improved, constraint conditions thus are no longer needed. exploited. An arbitrary type can embedded into base controller, new controller capable optimizing tuning parameters satisfying Lyapunov stability simultaneously. Simulation results verify effectiveness advantage our methods.