Neural network based robust dynamic inversion flight control
基于神经网络的鲁棒自适应逆飞行控制

An adaptive dynamic inversion controller design method based on neural network(NN) is proposed for super maneuverable fighter aircrafts.The basic control law is developed in nonlinear dynamic inversion;on-line learning neural networks are implemented to compensate inversion errors due to modeling error or actuator faults.Simulations of compensating simplified calculation and actuator failure are presented.The results demonstrate that through adaptively compensating inversion error by neural networks,the limitation of needing accurate mathematical model in dynamic inversion method is released and the robustness of the whole system is greatly enhanced.