多智能体系统的分布式采样一致性控制

项目名称： 多智能体系统的分布式采样一致性控制

项目编号： No.61273030

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： 肖峰

作者单位： 哈尔滨工业大学

项目金额： 80万元

中文摘要： 多智能体系统是目前现代控制理论中的重要研究对象之一，具有广泛的工程应用背景和重要的理论研究价值，得到了科研工作者的广泛关注。本项目将研究基于时间驱动和事件驱动控制策略的多智能体系统的采样一致性协调控制问题，包括大采样间隔条件下的周期和非周期采样一致性问题和基于局部信息的数据采样和控制器更新的事件检测与触发机制，并研究采样一致性理论框架下的多机器人交汇控制、队形控制及其连通性保持问题。利用控制理论、矩阵分析、代数图理论、概率论等基本理论工具，解决时变有向通信拓扑、时变时滞、异步个体动态等带来的协议设计与稳定性分析难点，丰富和发展采样一致性控制理论，为多机器人分布式协同提供有益的指导。

中文关键词： 多智能体系统；一致性问题；采样控制；事件驱动控制；边事件

英文摘要： In recent years, multi-agent systems have become an important research topic in modern control theory and attracted considerable attention from many researchers because of their theoretical challenges and potential applications. In this project, we will study sampled-data consensus problems in multi-agent systems based on the time-driven and event-driven control techniques, including the problem under periodic/non-periodic data sampling with large sampling periods and the presentation of local information based event-detecting and event-triggering rules for data sampling and controller updating; we will also study the rendezvous and formation problems and their connectivity preservation in multi-robot systems within the framework of sampled-data consensus. By using the tools of control theory, matrix analysis and probability theory, we will try to overcome the difficulties in protocol design and stability analysis, caused by time-varying directed interaction topologies, time-varying delays and asynchronous dynamics, and contribute to the development of sampled-data consensus theory.

英文关键词： Multi-agent systems；consensus problems；sampled-data control；event-driven control；edge events