中文摘要：

蜂拥行为的产生来源于对飞鸟、鱼群、蚂蚁以及蜜蜂行为的研究，这些群体通过每个个体的局部信息交互而产生统一的觅食、迁徙、归巢等行为。多智能体的自组织行为促进了蜂拥行为在工程上的应用，包括大量在环境中移动的传感器、媒介的平行同步传输、军事任务中的侦查、监视和战斗协作。主要研究内容是具有虚拟领导者的多智能体蜂拥控制算法，分别对群体中只有一部分智能体具有引导信息和虚拟领导者具有变化速度的情况作了分析。通过个体之间的局部感知作用和相应的反应行为，使系统整体呈现一定的协调行为。

英文摘要：

Flocking behavior of the production comes from the behaviors of birds, fish, ants and bees. Through exchanging local information with each individual, these groups produce unified behaviors of foraging, migration, homing, and other acts. Flocking is a large number of agents that interact with each other through a normal external command. The self - organizing of multi - agent develops the flocking behavior in engineering applications, including a large number of sensors mobile in the environment, the parallel and synchronous transmissions of media, and the reconnaissance, surveillance and combat collaboration in military tasks. In this paper, a single virtual leader of the multi - agent flocking control algorithm was studied. Its main targets are to study the case that only part agents have the inducting information and the cased that virtual leadership has the changing guiding paces. By means of the local perception of individuals and the corresponding reaction behaviors, the whol system presented a certain coordination