针对一类不确定非线性系统的跟踪控制问题,在考虑建模误差、参数不确定和外部干扰情况下,以良好的跟踪性能及强鲁棒性为目标,提出基于自组织小脑模型(self-organizing wavelet cerebellar model articulation controller, SOWCMAC)的鲁棒自适应积分末端(terminal)滑模控制策略。首先,将小脑模型、自组织神经网络和小波函数各自优势相结合,给出一种SOWCMAC,以保证干扰估计方法具有快速学习能力和更好的泛化能力。其次,设计两种改进的terminal滑模面构造方法,并分别给出各自的收敛时间。然后,基于SOWCMAC和改进的积分terminal滑模面,给出不确定非线性系统鲁棒自适应非奇异terminal控制器的设计过程,其中通过构造自适应鲁棒项抑制干扰估计误差对系统跟踪性能的影响,并利用Lyapunov理论证明闭环系统的稳定性。最后,将该方法应用于近空间飞行器姿态的控制仿真实验,结果表明所提出方法有效性。%We propose a robust adaptive integral terminal sliding mode control method using self-organizing wavelet cerebella model articulation controller (SOWCMAC) for a class of uncertain nonlinear systems with modeling error, parameter uncertainty and external disturbances to achieve the desired tracking performance and strong robustness. Firstly, we make use of the advantages of cerebella model articulation controller, self-organizing neural networks and wavelet function in developing the SOWCMAC to ensure the fast learning ability and desirable generalization ability. Next, we design two kinds of improved integral terminal sliding surfaces and express their convergence time in the analytical form. With the SOWCMAC and improved integral terminal sliding surfaces, we develop the robust adaptive nonsingular terminal controller for the uncertain nonlinear systems. The adaptive robust term can offset the impact of the approximation errors for the system. The stability of the closed-loop system is proved by using the Lyapunov theory. The method is applied to control the attitude system of a near space vehicle. The results show that the proposed method is effective.
展开▼
