信息传输不连续的多智能体系统的一致性分析与控制

项目名称： 信息传输不连续的多智能体系统的一致性分析与控制

项目编号： No.61203150

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 自动化学科

项目作者： 高彦平

作者单位： 北京工商大学

项目金额： 23万元

中文摘要： 由于在深空探测、军事协同侦查等方面的广泛应用，多智能体系统已经成为了一个热门的研究领域，而一致性是其研究中的基本问题之一。本课题采用图论、矩阵理论、稳定性理论、控制理论等工具研究信息传输不连续的多智能体系统的一致性，其中采样器的采样时间与保持器的更新时间可以不同，并寻求理论结果在多机器人协调控制领域中的应用。首先针对连续多智能体系统，设计一类一致性协议，并分析拓扑、时滞、离散时刻的分布、采样器的采样周期与保持器的更新周期的数量关系对一致性的影响。其次，针对二阶和高阶连续多智能体系统，在部分状态信息不可测量的情形，设计静态或动态输出反馈形式的协议，并建立一致性成立的判定条件。最后，设计一类有界的协议，分析时滞、拓扑等对一致性的影响。对信息传输不连续的多智能体系统一致性的研究，不仅可以促进一致性理论的发展，而且也为相关应用领域，尤其是多机器人协调控制领域，提供重要的理论基础。

中文关键词： 多智能体系统；一致性；信息传输不连续；时变拓扑；

英文摘要： Because of the wide applications in deep space exploration, military cooperative investigation, etc, multi-agent systems have been an active research field, and consensus is one of fundamental issues in the study of multi-agent systems. By applying graph theory, matrix theory, stability theory,control theory, etc, the consensus of multi-agent systems is studied in the case of intermittent interaction, where the sampling time and the update time may be different, and the applications of theoretical results are further addressed in coordinated control of multi-robots. First, a protocol is designed for continuous-time multi-agent systems, and it is analyzed that how topology, time-delay, the distribution of discrete times, and the quantitative relationship between sampling period and update period affect consensus. Second, in the case that partial state information can not be measured, a static or dynamic output feedback protocol is put for second-order or high-order continuous-time multi-agent systems, and some sufficient conditions are established for consensus. Third, a bounded protocol is designed, and the effect of time-delay, topology, etc, on consensus is also discussed. The study on the consensus of multi-agent systems with intermittent interaction can not only promote the development of consensus theory

英文关键词： multi-agent systems；consensus；intermittent interaction；time-varying topology；