连续/离散重复过程的鲁棒滤波研究

项目名称： 连续/离散重复过程的鲁棒滤波研究

项目编号： No.60804002

项目类型： 青年科学基金项目

立项/批准年度： 2009

项目学科： 金属学与金属工艺

项目作者： 吴立刚

作者单位： 哈尔滨工业大学

项目金额： 20万元

中文摘要： 重复过程是一类具有重要实际应用背景的特殊二维模型的重复控制系统，其在机器人控制、金属锻造以及迭代学习控制等领域中有着广泛而重要的应用。滤波在航空、航天、工业过程控制以及社会经济等领域具有非常广泛的应用。研究重复过程的鲁棒滤波是控制和信号处理领域的前沿研究课题，也是重复过程在实际工程应用中亟待解决的关键问题。本项目针对连续/离散重复过程，研究多种性能指标下鲁棒滤波器的设计方法，并研究鲁棒降阶滤波器的设计方法。在此基础上，进一步研究重复过程在参数存在随机跳变情况下的鲁棒滤波问题；研究在重复过程中存在传感器非线性和部分传感器失效等复杂环境下的鲁棒滤波问题。此外，本项目的部分理论研究成果将在空间机器人控制中进行尝试性应用研究。本项目的宗旨是提出一套较为完整的连续/离散重复过程的鲁棒滤波方法，为工程技术人员提供较实用的鲁棒滤波器的设计方法。基于此项目发表论文39篇，SCI收录33篇,EI收录35篇。其中，3篇论文发表在Automatica上面，8篇论文发表在IEEE Transactions汇刊上。

中文关键词： 线性重复过程；鲁棒滤波；状态估计；参数跳变

英文摘要： Linear repetitive processes are a distinct class of 2-D systems, which have extensive applications in the practical industry, such as robot control, long-wall coal cutting and metal rolling operations. On the other hand, filtering problem has received great attention due to the fact that it has important applications in the areas of aerospace, astronautics, industrial process and social economics.The problem of robust filtering for linear repetitive processes is a front reasearch topic in control and signal processing fields, and is also a key step to push the linear repetitive processes into the practical applications. This project aims at the key problems in robust filtering area, and investigates the multi-objective filtering problem and reduced-order filter design approaches. Moreover,this project also investigates the robust filtering problem for linear repetitive processes with parameter jumping and under complex environment such as systems with sensor nonlinearity, partial sensor invalidation and limited communications. Part of the theoretical results of this project will be tentatively applied to the control and state estimation of aerospace robots. The main aim of this project is to propose a series of robust filtering methods for discrete- differential linear repetitive processes, and provide useful design methods and theory for practical engineers. Based on this project, 39 papers are published, in which 33 papers are indexed by SCI and 35 papers are indexed by EI, and 3 papers are published in Automatica and 8 papers are published in the journal of IEEE Transactions.

英文关键词： Linear repetitive processes; robust filtering; state estimation; parameter jumping