迭代学习控制
离散时间和连续时间
计算机科学
控制(管理)
控制理论(社会学)
线性系统
数学
应用数学
人工智能
统计
数学分析
作者
Ying Kong,Xiao‐Dong Li,Xuefang Li
标识
DOI:10.1016/j.jfranklin.2024.106702
摘要
Different from the existing iterative learning control (ILC) works for dynamical systems with iteration-varying trial lengths, in which the terminal time points are not identical, this paper presents an ILC approach for linear discrete time-varying (LDTV) systems with different initial time points. A piecewise ILC law is designed by using the segment compensation techniques in time axis. Through a rigorous mathematical analysis, it is theoretically proved that as the mathematical expectation of the iterative initial state equals to the desired initial state, the ILC tracking error at the maximum universal time interval can be driven into a bounded region in mathematical expectation sense. Particularly, the mathematical expectation of the ILC tracking error at the maximum universal time interval can even be driven to zero under a certain initial state condition. Numerical simulations are provided to verify the effectiveness of the proposed ILC scheme.
科研通智能强力驱动
Strongly Powered by AbleSci AI