基于RCMAC干扰观测器的高超声速飞行控制

利用自回归小脑模型神经网络(recurrent cerebella model neural network, RCMAC)良好的非线性逼近能力和自学习能力，结合反馈线性化和反演控制方法，提出了一种自适应非线性控制策略，用于高速再入飞行器控制系统的设计。该方案将RCMAC干扰观测器(recurrent cerebella disturbance observer, RCDO)用于估计系统模型的不确定项，同时采用反演控制方式设计伪线性控制项，并利用符号函数逼近误差的上界，根据Lyapunov稳定性理论设计了权值更新规则，保证闭环系统信号有界。高速再入飞行器的六自由度仿真结果验证了方法的有效性和鲁棒性。

Nonlinear control for hypersonic vehicles based on RCMAC disturbance observer

In virtue of the nonlinearity approximating ability and self learning ability of recurrent cerebella model neural network (RCMAC), an adaptive nonlinear control strategy combined with feedback linearization and backstepping method is established and adopted to design controller of high-speed reentry vehicle. An RCMAC disturbance observer (RCDO) is developed to deal with uncertainties in the system, and the backstepping method is utilized to design pseudo linear terms where signal functions are presented to estimate the upper bound of  approximation errors. The weights adjusting law is derived according to Lyapunov stability theory, which can guarantee the boundedness of all signals in the system. Six degrees of freedom simulation results demonstrate the effectiveness and robustness of the proposed approach.