汕头大学数学系

学术报告

当前位置: 首页 >> 科学研究 >> 学术报告 >> 正文
Convergence of Newton's Method for Nonlinear Deterministic Optimal Control Problems
日期: 2025-10-31      信息来源:      点击数:

走向现代数学学术报告 - 张海森教授(No. 848)

报告题目:Convergence of Newton's Method for Nonlinear Deterministic Optimal Control Problems

报告时间:2025年11月4日 11:00

腾讯会议ID:353 698 792

报 告 人:张海森 教授(四川师范大学)

邀 请 人:段月亮 博士

报告摘要:In this talk, we will discuss the convergence of Newton's algorithm for general nonlinear deterministic optimal control problems. Here, the control system is a deterministic nonlinear ordinary differential equation. We first proved the second-order differentiability of the cost functional under appropriate conditions and provided an accurate characterization of the second-order derivative of the cost functional with respect to the control in L^2 space. The local quadratic convergence rate of the Newton's algorithm is proved under a proper second-order sufficient condition for optimal controls in L^2 space. Compared to previous work, the second-order sufficient conditions in L^2 space are weaker and easier to verify.

报告人简介:张海森,四川师范大学教授,博士生导师,霍英东青年教师奖获得者。主要研究领域为随机优化与控制,在 SIAM Review, Trans. Amer. Math. Soc.,SIAM J. Control Optim.,J. Differential Equations等期刊发表学术论文近二十篇。主持国家自然科学基金2项,省重点项目1项,参与国自科基金重点项目1项。现担任SCI 学术期刊Mathematical Control and Related Fields编委。              

广东省汕头市翠峰路5号,汕头大学数学系 515821,Email:math@stu.edu.cn

Copyright 2003-2023 汕头大学数学系