回顾了自适应飞行控制技术、反馈线性化和模型逆理论 ,分析了误差动力特性 ,设计了自适应神经网络姿态控制系统。其中 ,模型逆基于悬停状态 ,基于神经网络的自适应控制律能够确保跟踪误差和控制信号的有界。仿真结果表明 :模型逆增强的非线性神经网络能够对无人直升机的不确定性和建模误差进行自适应。而且对 PD控制器和鲁棒项系数变化的仿真结果进行了比较 The adaptive flight control technique, feedback linearization and model inverse theory are reviewed. The error dynamic characteristics are analyzed and an adaptive neural network attitude control system is designed. Among them, the model inverse based on the hovering state, adaptive control law based on neural network to ensure that the tracking error and control signal bounded. The simulation results show that the model inversely enhanced nonlinear neural network can adapt to the uncertainty and modeling error of the unmanned helicopter. Moreover, the simulation results of PD controller and robust coefficient changes are compared