Performance Enhancement of Multiple Model Adaptive Control by Using Neural Networks

Multiple linear and BP NN (back propagation neural network) models are used to approximate the complex nonlinear system, and different model reference adaptive controllers based on these models and different switching mechanisms are applied to a nonlinear system to trace a reference trajectory. From the simulation, it can be shown that the multiple model adaptive control method proposed in this paper can improve the control performance greatly compared with conventional adaptive neural network controller.