项目名称: 基于马氏跳变的非线性多智能体的网络随机建模与协同控制
项目编号: No.61273027
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 自动化技术、计算机技术
项目作者: 王迎春
作者单位: 东北大学
项目金额: 80万元
中文摘要: 本项目申请基于马氏跳变的非线性系统网络控制的连续资助。在前一项目“基于马氏跳变理论的非线性随机网络控制系统的建模、分析与综合”基础上,这次主要针对多智能体的网络控制采用马氏跳变理论进行建模和协同控制,将随机网络控制理论从单一非线性系统(单智能体系统)扩展到多智能体的分析与协同控制。前一个项目中解决了3个关键问题,正研究最后一个问题。目前接受和发表论文15篇,其中SCI源期刊论文7篇,专著1部。本申请包括:1. 多智能体信息流在网络诱导时延和丢包同时存在情况下,研究基于马氏跳变理论的复杂多智能体的随机网络拓扑结构和协同控制问题;2.研究部分马氏跳变概率转移矩阵未知情况下多智能体的网络协同建模和控制;3.在传感/通讯信息异步更新情况下,基于马氏跳变理论的高阶多智能体的时滞相关的协同控制;4.针对复杂多智能体建模困难情况,采用模糊系统和随机马氏跳变理论相结合思想进行协同控制建模和模糊协同控制。
中文关键词: 网络控制;协同控制;多智能体系统;马氏跳变;模糊控制
英文摘要: In this project, we propose the continue support on the Markovian jumping based networked cooperative control for complex multi-agents. According to the front project on “Markovian jumping based model, analysis and synthesis of nonlinear stochastic networked control systems”, the main work is the networked cooperative control of multi-agents. The Markovian jumping theory based stochastic networked control is extent to the cooperative analysis and synthesis of multi-agents from single nonlinear system (single agent). In front project, three key problems have been achieved and remain the last problem. Now, 15 papers are accepted or published, in which 7 is in SCI source. One monograph is published as Editor. The main novelties include four: 1) In sensing/communication network dropout and delay, Markovian jumping stochastic theory based cooperative modeling and control of multi-agents; 2) In partly unknown Markovian jumping matrix, the networked cooperative model and cooperative control of multi-agents; 3) In asynchronous samplings/communication, the Markovian jumping based delay-dependent cooperative control of high order multi-agents systems; 4) In case of the difficult model, the fuzzy cooperative control of multi-agents by combining the fuzzy model approach and Markovian jumping theory.
英文关键词: networked control;cooperative control;multi-agent systems;Markovian jump;fuzzy control