时延和拓扑时变的高阶多智能体系统编队-合围控制方法研究

Theories on formation-containment control of high-order multi-agent systems have potential applications in the formation attack of multiple manned/unmanned combat aerial vehicles and cooperative penetration of multiple missiles, etc. Time-varying delays and topology switching are inevitable in practical multi-agent systems, and the effects of them on the formation-containment control of multi-agent systems are unnegligible.. This research project intends to investigate the analysis and design problems for high-order multi-agent systems with both time-varying delays and switching directed topologies. First of all, time-varying formation control problems for high-order multi-agent systems with time-varying delays and switching directed topologies will be dealt with. Sufficient/necessary and sufficient conditions for multi-agent systems to achieve time-varying formations and formation feasibility will be proposed. An explicit expression of the formation reference function and approaches to specify the motion modes of the formation reference will be presented. Approaches to design the time-varying formation protocol and expand the feasible formation set will be given. Then containment control problems for high-order multi-agent systems with time-varying delays and switching directed topologies will be studied. Criteria for multi-agent systems to achieve containment and approaches to determine the gain matrices in the containment protocol will be proposed. Finally, formation-containment control problems for high-order multi-agent systems will be investigated. Sufficient conditions for high-order multi-agent systems with time-varying delays and switching directed topologies to achieve formation-containment will be presented, and approaches to design the formation-containment protocol will be proposed.

高阶多智能体系统编队-合围控制理论在有人/无人战斗机混合编队协同攻击及多导弹协同突防等领域具有潜在的应用价值。时变时延和拓扑切换在实际多智能体系统中不可避免，对编队-合围控制的影响不容忽视。. 本项目拟研究时变时延和有向切换拓扑同时存在条件下高阶多智能体系统编队-合围控制的分析和设计问题。首先，研究多智能体系统在上述条件下的时变编队控制问题，给出多智能体系统实现时变编队及编队可行的充分/充要条件，给出状态编队参考函数的显式表达式以及对编队参考函数运动模态进行配置的方法，提出对编队协议进行设计的方法以及扩展可行编队集合的方法。其次，研究多智能体系统合围控制问题，给出多智能体系统在上述条件下实现合围的判据以及确定合围协议中增益矩阵的方法。最后，研究多智能体系统编队-合围控制问题，给出多智能体系统在上述条件下实现状态编队-合围的充分条件和对编队-合围控制协议进行设计的方法。

分布式编队-合围控制是多智能体系统智能涌现的保障，更是军事领域技术制高点。本项目研究了时延和拓扑时变条件下的多智能体系统编队-合围控制分析与设计中的难点及关键问题，提出了时变编队可行性及编队参考函数的分析方法，解决了分布式编队控制中队形时变的难题；设计了在时变通信时延与拓扑切换条件下的时变编队控制器，给出了多智能体系统实现时变编队的判据、确定了通信时延和拓扑切换对编队可行域和编队参考函数的影响；构造了具有时变通信时延与拓扑切换的合围控制协议，给出了高阶多智能体系统在通信约束条件下实现合围的充分条件以及对控制器的分布式设计算法；提出了编队-合围控制模型和方法，突破了分层耦合协同的难题，实现了通信时延、输出反馈等条件下的编队-合围控制；拓展研究了时变编队跟踪控制问题，解决了具有领导者未知时变输入以及多个领导者等条件下的编队跟踪难题；将编队合围-控制理论应用到导弹和无人机等实际对象上，并在自主研制的多无人机平台上进行了系列飞行试验，验证了相关理论的有效性和实用性，推动了协同控制理论走向应用。在该项目的资助下，共发表(含录用)学术论文93篇，其中SCI及Trans.汇刊论文39篇，EI论文54篇，包括IEEE汇刊论文12篇(长文10篇)，1%ESI高被引论文2篇。项目负责人董希旺获得“吴文俊人工智能科学技术奖”优秀青年奖，并入选中国科协青年人才托举工程。