Simulation Analysis of the Neural Network Based Missile Adaptive Control with Respect to the Model Uncertainty

신경회로망 기반 미사일 적응제어기의 모델 불확실 상황에 대한 시뮬레이션 연구

This paper presents the design of a neural network based adaptive control for missile. Acceleration of missile by tail fin control cannot be controllable by DMI (Dynamic Model Inversion) directly because it is non-minimum phase system. To avoid the non-minimum phase system, dynamic model inversion is applied with output-redefinition method. In order to evaluate performance of the suggested controllers we selected the three cases such as control surface fail, control surface loss and wing loss for model uncertainty. The corresponding aerodynamic databases to the failure cases were calculated by using the Missile DATACOM. Using a high fidelity 6DOF simulation program of the missile the performance was evaluates.