基于BP网络的BTT导弹逆控制方法

摘要

A BP neural network applying Levenberg-Marquard lgorithm was used to realize the identification of the inverse dynamic system model of missile. And the system which makes the indentificaiton model and the BTT missle control system are in series was dynamic pseudolinear. Then a method of on-line self-learning control strategy which combines inverse control based on classical control with neural network,was proposed in the system. It is proved theoretically that the method can realize the missle three channel' s decoupling control and make tract indistinctively and gradually. The results of simulation show that in the design of BTT missle control system,neural network inversion control method can realize nonlinear control,the control sheme is feaisble and meet the designing requiremnet.%运用BP神经网络结构和Levenberg-Marquardt算法,对导弹动力学特性进行动态逆模型辨识,并以辨识模型为控制器与BTT导弹控制系统串联构成动态伪线性系统,进而应用逆系统方法设计了一种用于解决BTT导弹非线性控制问题的经典控制与神经网络在线自学习控制相结合的控制方法.从理论上证明了该控制方案可以实现对导弹三通道的解耦控制,输出渐进无差跟踪.仿真结果表明,在BTT导弹控制系统设计中,神经网络逆系统方法完全可以实现非线性控制,满足设计要求.