The Anti-windup Adaptive Control of Aircraft Based on Neural Network

According to the unfitness flight control of conventional adaptive control on aircraft actuator with magnitude saturation and rate saturation,an anti-windup adaptive control method of aircraft based on neural network and pseudo-control hedging is proposed.Firstly,regarding the model reference adaptive control theory and the combination of dynamic inversion and neural network adaptive control method,the characteristics of model tracking error dynamics is analyzed,and the online neural network is designed to compensate aircraft dynamic inversion error such as the un-modeled error and parameter uncertainty.And then,with the consideration of actuator saturation,the isolation of adaptive element and actuator saturation is achieved by pseudo-control hedging to gain excellent control of aircraft at each channel.Finally,the neural network compensation for dynamic inversion error and pseudo-control hedging anti-windup is validated by the UH-60 Black Hawk helicopter.The simulation results show the effectiveness of the proposed method.