论文部分内容阅读
针对任意初态情形,借助于初始修正吸引子的概念,讨论不确定时变系统能够达到实际完全跟踪性能的迭代学习控制方法.闭环系统中含有限时间控制作用,在预先指定的区间上实现零误差跟踪,且起始段的系统输出轨迹也可预先规划.分别讨论部分限幅学习与完全限幅学习,证明闭环系统中各变量的一致有界性以及误差序列的一致收敛性.变量有界性证明得益于提出的限幅学习算法,特别是完全限幅学习算法可确保参数估值的变化范围.
In this paper, an iterative learning control method based on initial modified attractor is proposed to deal with an arbitrary initial state, and an iterative learning control method is proposed to solve the problem that an uncertain time-varying system can achieve full tracking performance. The closed-loop system includes a finite-time control function, Error tracking, and the initial output of the system trajectory can also be pre-planned part of the study discussed the full limiting amplitude learning and clipping, the closed-loop system to prove the uniform bound of the variables and the error sequence of the uniform convergence of variable bound The proof of sexuality is due to the proposed limiter learning algorithm, especially the full-limiting learning algorithm, which ensures a wide range of parameter estimates.