报告摘要：In this talk, the problem of domain aiming control is formulated for stochastic controllable systems. As preliminary results, a recurrence criterion of a domain and an upper bound to the mean of the first hitting time for the domain are presented for general stochastic differential equations. Under certain conditions, it is shown that a stochastic system is recurrent with respect to a domain, and the first hitting time has a density function and its distribution function and Laplace transform can be expressed as solutions of Dirichlet problems. Hence, estimates of distribution function and density function for the first hitting time can be derived in terms of numerical methods for solving associated PDEs. Finally, the problem of domain aiming control is considered for linear controllable stochastic systems. Controller design methods and an illustrative example are given.

报告时间：4月12日10：30 -11：10

报告地点：主楼二号会议室

报告人：尹居良，广州大学教授，博士生导师

个人简介：

尹居良，2003年博士毕业于中山大学概率论与数理统计专业，统计学专业博士生导师。入选省级人才计划。现任中国现场统计研究会常务理事、广东省高校统计学教学指导委员会副主任和中国工业与应用数学学会系统与控制数学专业委员会委员等职务。主要从事随机分析与随机控制、数理金融与保险精算、随机过程统计等领域的教学与研究工作。先后主持承担国家自然科学基金项目3项、省部级科研项目7项，正式发表各类学术论文60余篇，其中被SCI和SSCI索引收录42篇。2007年和2008年分别获广东省统计科学研究成果一等奖和全国优秀统计成果三等奖。