报告题目: ILC with Non-Repetitive Reference Trajectories, Iteration Varying Trial Lengths, and Asymmetric Output Constraints

报告人: 金旭

主持人: 陈强

报告时间: 2024年8月15日  15:00-16:00

报告地点: 信息楼D533

报告摘要：

    In this talk, we discuss a novel iterative learning control (ILC) scheme for non-repetitive reference trajectories tracking problems of robot manipulators over an iteration domain with varying trial lengths, subject to asymmetric constraint requirements on joint angles. To address iteration varying trial lengths, unlike the existing approaches based on the contraction mapping analysis, a new structure of ILC laws has been presented in this work, using analysis based on composite energy functions. A novel universal barrier function is proposed to deal with joint angle constraints. We show that under the proposed novel ILC scheme, beyond a small initial time interval in each iteration, the joint angle tracking error is uniformly converging to zero over the iteration domain, and the joint velocity tracking error is asymptotically converging to zero in the sense of certain L2 norm. In the end, a simulation example on a two-degree-of-freedom robot manipulator is presented to demonstrate the efficacy of the proposed scheme.

报告人简介：

    金旭博士，美国肯塔基大学助理教授、博士生导师。2013年获得新加坡国立大学电子计算机专业一等荣誉学士学位，2015年获得加拿大多伦多大学电子计算机专业硕士学位，2018年获得美国佐治亚理工大学数学硕士学位，2019年获得美国佐治亚理工大学航天工程博士学位。2019年至今在美国肯塔基大学机械工程系工作。发表论文70余篇，引用量3900余次。获2024年度美国国家自然科学基金（NSF）CAREER奖项。独立主持美国国家自然科学基金（NSF）项目两项，并多次担任美国国家自然科学基金委（NSF）评审专家。另主持美国航空航天局（NASA）州级项目一项。金博士在全球前2%各领域科学家榜单的斯坦福“2022年度科学影响力排行榜”总排名7562。主要方向包括智能控制理论、非线性多智能体协同、非线性系统受限问题等研究。