Design of synchronising controllers

自主智能体的同步任务可以通过一个网络控制器来解决，该控制器引导智能体朝着共同的轨迹前进。在大量关于多智能体系统同步的文献中，通常研究相同智能体的渐近同步。该项目将同步视为一个控制问题，其中除了渐近同步的要求外，还规定了对整个系统的暂态行为和干扰抑制的要求。在支撑的第一阶段，发展了具有相同动力学的智能体的同步设计方法。网络控制器作为优化问题的解得到，其中目标函数的选取反映了智能体的暂态行为。在本请求提供支助的这段时间内，必须制定使代理人与个人动态同步的方法。对于代理数量不断变化的多代理系统以及具有基于事件的通信的系统的同步控制器的设计，必须对这些方法进行修改。该设计方法将在仿真和实验中得到验证，并应用于移动机器人的编队。

The task of synchronizing autonomous agents can be solved by a networked controller that steers the agents towards a common trajectory. In the extensive literature on synchronization of multiagent systems usually asymptotic synchronization of identical agents is investigated. This projectconsiders synchronization as a control problem, where in addition to the requirement on asymptotic synchronization, requirements on the transient behavior and disturbance rejection of the overall system are specified. In the first time period of support design methods for the synchronization ofagents with identical dynamics have been developed. The networked controller was obtained as the solution of an optimization problem, where the objective function was chosen so as to reflect the transient behavior of the agents. In the time period of the present request for support, methods forthe synchronization of agents with individual dynamics have to be developed. These methods have to be modified for the design of synchronizing controllers for multi-agent systems with a changing number of agents as well as for systems with event-based communication. The design methodswill be approved in simulations and experiments by application to a formation of mobile robots.