组合连通拓扑下基于事件触发的多智能体快速一致性算法

Event-triggered Fast Consensus Algorithm for Multi-agent Systems Under Jointly-connected Topology

针对组合连通拓扑下多智能体系统控制过程中存在通信和计算资源损耗大以及系统收敛速度慢等问题,提出一种新的具有状态预测器的事件触发一致性控制协议,通过设计状态预测器使每个智能体都能对其邻居智能体的未来状态作出预测;同时,对于智能个体给出了基于状态信息的事件触发条件,当状态误差满足该条件才触发事件.在该控制策略下多智能体系统可在节约通信和计算资源的同时具有更快的收敛速度.利用Lyapunov稳定性理论和代数图论,证明了所提事件触发控制策略能够有效实现组合连通拓扑结构下的平均一致性,且不存在Zeno行为.仿真实例进一步验证了理论结果的有效性.

This paper investigates the event-triggered consensus problem of multi-agent systems under jointly-connected topology. In order to reduce the unnecessary waste of limited communication and computing resources as well as improve convergence rate, a novel event-triggered consensus control law with state predictor is proposed. In particular, every agent can predict the future state of its neighbor; meanwhile, for each agent a state-dependent event condition is given, and only when state error satisfies this event condition, can the event be triggered. This control strategy can lead to a significant reduction of information communication burden in a multi-agent network and improvement of convergence rate. Based on the Lyapunov stability theorem and algebraic graph theory, the proposed event-triggered control strategy is proven to be able to implement the average consensus when the topology is jointly-connected. Moreover, such strategies can exclude Zeno-behavior. Finally, numerical simulations are given to illustrate the effectiveness of the theoretical results.