有界输入下基于backstepping的迭代学习控制

Iterative learning control with bound input based on backstepping

讨论一类含饱和执行机构的非匹配不确定非线性系统的输出跟踪问题.结合反演(backstep-ping)设计方法和迭代学习控制,提出一种输入限幅下的backstepping迭代学习控制方案.学习控制用于学习周期性的系统不确定性,backstepping方法用于处理非匹配不确定性,并且利用低通滤波器削弱饱和执行机构的影响.通过Lyapunov方法,证明所设计的控制器不仅可保证闭环系统中所有信号一致有界,而且随迭代次数增加,跟踪误差收敛到零.仿真结果表明控制器的有效性和可行性.

In this paper, an output tracking problem is addressed for a class of nonlinear mismatched uncertain systems with saturating actuators. A backstepping iterative learning control law with input saturation is developed by integrating backstepping method and iterative learning control. The learning control is utilized to learn the periodic uncertainties and the backstepping method is to deal with the mismatched uncertainties. A low pass filter （LPF） is utilized to alleviate the effect of saturating actuators. By using Lyapunov method, the derived controller guarantees all signals of the closed-loop systems are bounded. The tracking error will converge to zero as iteration increases. Finally, a designed example is presented and the result demonstrates the controller＇s effectiveness and feasibilities.