输入限幅下的迭代学习控制  

Iterative learning control with input saturation

作　　者：金奎[1] 孙明轩[1] 

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310023

出　　处：《浙江工业大学学报》2011年第5期579-585,共7页Journal of Zhejiang University of Technology

基　　金：国家自然科学基金资助项目(60474041;60874041)

摘　　要：基于Lyapunov-like方法,提出在有限作业区间上实现跟踪的带限幅的迭代学习控制器设计.针对含有非参数化及部分参数化不确定性的非线性系统,设计鲁棒迭代学习控制器,但该方法不要求已知不确定特性的范数界.针对参数化不确定非线性系统,设计自适应迭代学习控制器.在系统满足有界输入有界状态(BIBS)假设条件及控制输入饱和限幅作用下,统一对上述三类系统进行分析.理论分析结果表明:该方案可保证闭环系统中所有变量的有界性,并且跟踪误差能够在整个作业区间上收敛到零,实现完全跟踪.仿真验证了该算法的有效性.Based on Lyapunov-like approach, an iterative learning controller with input saturaiton for tracking in the limited time interval is proposed. For a class of nonlinear systems with non- parametric and partially parametric uncertainties, a robust iterative learning controller is designed. But this method does not require the known of the uncertain norm bounds. An adaptive iterator learning controller is designed for the nonlinear system with parametric uncertainties. The three kinds of systems above are analyzed under the systems to meet the bounded input bounded state （BIBS） assumptions and limiting the role of control input saturation. The results show that the proposed learning algorithm can guarantee boundedness of all variables in closed- loop system. The tracking error can be guaranteed to converge to zero over entire time interval and perfect tracking can be achieved. The simulation result shows the algorithm is effective.

关 键 词：迭代学习控制 输入限幅 有界输入有界状态 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]