Abstarct: This study presents a new approach to model predictive control baxsed on new identification method by using neural identifiers with tapped delay lines added to the input laxyer of the artificial neural network system. The identifier uses the back propagation method in order to minimize the output error. For neumerical simulation first by considers an UAV as a controllable model, tries to investigate several articles. Finally, by using system identification methods (for system modeling) with a neural network, presents a precise model of the system and examine its result by applying a model baxsed predictive control.