符号图下多智能体系统事件驱动二分一致性

Event-triggered Bipartite Consensus for Multi-agent Systems Associated with Signed Graphs

  • 摘要: 为减少对网络带宽、通信成本及智能体自身能量等有限资源的不必要消耗,研究了无向符号图拓扑结构下事件驱动的一阶多智能体系统二分一致性问题.根据符号图的特性,设计了协作和竞争交互关系并存的二分一致性控制协议,使得多智能体系统收敛到两个模值相同但符号不同的状态;利用代数图论和矩阵论,分析并得到了多智能体系统实现二分一致性的充分条件,同时得到智能体最终一致性值.在二分一致性研究中,典型的控制方法是假定周期采样控制,这将会造成能耗增加、带宽占用高等问题.引入事件驱动控制,即构建事件驱动函数,当某个智能体状态测量偏差满足驱动函数时,该智能体触发事件,且所有智能体更新控制输入.智能体之间无需连续通信,实现降低能耗、提高资源利用率的目的.最后,仿真验证了所提理论算法的有效性.

     

    Abstract: To reduce the unnecessary waste of limited resources, such as network bandwidth, communication cost, and the energy of the agent itself, the event-triggered problem of bipartite consensus for multi-agent systems with first order under the undirected signed graph topology is studied. On the basis of the properties of the connected signed graph, a novel bipartite consensus protocol with the coexistence of cooperative and competitive interactions is designed, which makes the interested states of each agent reach a consensus with identical magnitude but opposite sign. Furthermore, using the algebraic graph and matrix theories, the appropriate conditions for multi-agent systems to achieve bipartite consensus are obtained and the final agreement of each agent is achieved. In the study of bipartite consensus, the typical control method assumes periodic sampling control, which will cause problems, such as increased energy consumption and high bandwidth occupation. Event-triggered control is introduced, that is, an event-triggered function is constructed. When an agent state measurement error satisfies the function, the agent triggers an event. Then, all agents trigger the event and update the control input. Continuous communication between agents is not required, which reduces energy consumption and improves resource utilization. Finally, simulations are conducted to illustrate the efficiency of the theoretical results of the proposed method.

     

/

返回文章
返回