When Is A Nonlinear System Semiglobally Asymptotically Stabilizable by Sampled-Data Feedback?

主题:When Is A Nonlinear System Semiglobally Asymptotically Stabilizable by Sampled-Data Feedback?

主讲人:林威

时间:2025-11-26 10:00

地点:2号学院楼2202

组织单位:信息与智能科学学院

报告人简介

林威教授曾获得美国国家科学基金会(NSF)CAREER Award、Warren E. Rupp讲席教授、Robert Herbold Faculty Fellow Award,并被授予IFAC Fellow与IEEE Fellow等荣誉。他曾担任《IEEE Transactions on Automatic Control》(1999–2002)副编辑、《Automatica》(2003–2005)副编辑,《TAC》“New Directions in Nonlinear Control”(2003)特刊客座编辑,《International Journal of Robust and Nonlinear Control》(2005–2010)专题编辑,《Journal of Control Theory and Applications》(2005–2008)副编辑。他还曾任 IEEE 控制系统协会理事会成员(2003–2005),并担任2001年IEEE CDC(短文)与2002年IEEE CDC(特邀论文)的副程序主席。 此外,他曾多次受邀在IFAC与IEEE等国际会议上作大会报告,包括2004年在德国斯图加特举办的IFAC NOLCOS’04,以及2007年在比勒陀利亚举办的第七届IFAC非线性控制专题研讨会。


报告摘要

Most of the existing results in the literature were devoted to practical stabilization of nonlinear systemsby sampled-data feedback and very few resuits are available on asymptotic stabilization, due to thenature of “sample and hold and the hybrid closed-loop system. In this talk, we present recentbreakthrough in semiglobal asymptotic control of general nonlinear systems by sampled-datafeedback. For MlMO nonlinear systems, we prove that while global asymptotic stabilizability (GAS) maynot ensure either global or semiglobal asymptotic stabilizability (SGAS)by sampled-data statefeedback, global asymptotic local exponential stabilizability, (GALES) does imply the existence of a SGASsampled-data state feedback controller. Based on this state feedback result, we further present themain contribution of this talk: GALES and uniform observability imply semiglobal asymptoticstabilizability, by sampled-data output feedback, which is indeed a hybrid version of the separationprinciple for nonlinear systems. The development of the digital separation principle is made possibleby employing a subtle Lyapunov arqument that makes the estimation of domains of attraction andsemiglobal asymptotic analysis intuitive and simple, without involving intricate Lyapunoy functions andthe corresponding level sets. Examples and results on SGAS by sampled-data feedback forrepresentative classes of nonlinear systems are also given as illustrations.