有输入未建模动态和不确定性导弹控制器设计

Robust Adaptive Controller Design for a Missile System with Input Unmodeled Dynamics and Uncertainties

在考虑导弹控制系统同时存在输入未建模动态和参数不确定性的情况下,提出了一种基于RBF神经网络和反演技术的鲁棒自适应控制器的设计方法。首先用两个RBF神经网络对未建模动态进行了补偿,然后利用反演技术设计了鲁棒自适应控制器,对不确定性进行了处理,并用Lyapunov稳定性理论推导出RBF神经网络的权重矩阵调节率以及相关的自适应律,证明了系统的全局稳定性,最后通过仿真计算验证了设计的正确性。

When considering a missile system robust adaptive controller based on RBF neural with both input unmodeled dynamics and uncertainties, a networks and backstepping is proposed. Firstly, we use two RBF neural networks to approximate the unmodeled dynamics and its inverse. And the robust adaptive controller based on backstepping is presented, secondly, we attain the tuning law of RBF neural networks and corresponding adaptive control law through Lyapunov stability theory, verifying the global stability of the system. At the end of this paper, the corresponding 6DOF simulation of the missile is given to validate the correctness of the design method.