基于滑模观测器的非线性系统迭代学习控制

Iterative learning control for nonlinear system based on sliding mode observer

针对相当广泛的一类非线性系统有限时间轨迹跟踪问题,提出了一种基于滑模观测器的迭代学习控制算法。根据系统的非线性特性,利用一种滑模观测器对系统的状态进行估计,根据估计信号设计了一种类D型开环迭代学习控制律。这种控制方法不需要对系统的跟踪误差信号进行微分,从而对系统的量测噪声不敏感。给出了控制算法的收敛性证明,通过仿真实验证明了这种算法的有效性。

A sliding mode observer based iterative learning control scheme for a kind of nonlinear system to track a trajectory with a finite time interval is presented. According to the nonlinear characters of the system, the states are estimated by a sliding mode observer, and the iterative learning control law is designed using the estimate signals. The method is not hypersensitive with the measurement error since without using the derivate of the tracking error signal in the control law. The convergence of the proposed control law is proved, and numerical simulation results show the validation of the control law.