论文部分内容阅读
针对飞行控制系统中状态向量不完全可测量的问题,设计了一种高增益神经网络自适应观测器.在常规高增益观测器的基础上引入径向基(RBF,Radial Basis Function)网络自适应项,对建模误差和外界干扰进行在线估计.将高增益观测器与基于动态面的反步控制相结合,提出了一种神经网络自适应反步控制方法.引入一阶滤波器,避免了传统反步控制中的“计算膨胀”问题.基于Lyapunov稳定性理论,给出自适应输出反馈控制器和RBF网络权值向量的自适应律,并证明了闭环系统是半全局一致有界.从航迹角控制系统仿真结果可以看出,航迹角能够较好地跟踪指令信号,不受建模误差和外界干扰的影响,所设计的观测器具有良好的收敛性,控制系统具有较高的鲁棒性.
Aiming at the incomplete measurable state vector in flight control system, a high-gain neural network adaptive observer is designed. Based on conventional high-gain observer, Radial Basis Function (RBF) network adaptive Item, the modeling error and the external disturbance are estimated on-line.The high gain observer is combined with the backstepping control based on dynamic surface to propose a neural network adaptive backstepping control method.First-order filter is introduced to avoid In the traditional backstepping control, the problem of “calculating expansion ” is given. Based on the Lyapunov stability theory, an adaptive law of adaptive output feedback controller and weight vector of RBF network is given, and the closed-loop system is proved to be semi-globally uniformly bounded. It can be seen from the simulation results of the track angle control system that the track angle can track the command signal well without modeling error and outside interference. The designed observer has good convergence and the control system has higher Robustness.