导弹大迎角飞行时,系统非线性特性非常明显,各通道间有很严重的气动交叉耦合现象。为实现对系统的非线性解耦,构造了基于神经网络动态逆的大迎角导弹解耦控制器,设计了非线性动态逆系统,利用RBF神经网络逼近逆误差。仿真实验结果表明:所设计的控制系统具有良好的解耦和指令跟踪能力。 When the missile is flying at high angle of attack, the nonlinearity of the system is very obvious. There is a serious phenomenon of pneumatic cross-coupling between the channels. In order to realize the nonlinear decoupling of the system, a decoupling controller of missile with high attack angle based on neural network dynamic inversions was designed and a nonlinear dynamic inverse system was designed. The RBF neural network was used to approximate the inverse error. The simulation results show that the designed control system has good decoupling and command tracking ability.