Distributed Encirclement Points Deployment of Autonomous Mobile Multi-Agents Based on Finite-Time Consensus

基于有限时间一致性的自主移动多智能体分布式合围点部署

School of Astronautics, Northwestern Polytechnical University, Xi'an 710072, China

Received: 10 January 2020

Abstract

In the many-to-one pursuit-evasion game, autonomous mobile multi-agents can effectively improve the probability of successful hunting by achieving the cooperative encirclement to the target from multiple directions. To solve the encirclement points deployment, a distributed control protocol is proposed based on finite-time consensus. Firstly, the encirclement points deployment is formulated as the coverage control problem on a circle. The presented distributed control protocol can realize uniform coverage on the circle, i.e., optimal encirclement around the target. Secondly, by introducing the strategy of time-varying feedback gain, the convergence rate of closed-loop system turns into finite-time consensus from asymptotical consensus, such that the efficiency of the deployment is improved greatly. Finally, numerical simulations are carried out to verify the effectiveness of the presented distributed control protocols.

摘要

在多对一的追逃任务中，自主移动多智能体从多个方向实现对目标的协同合围，可以有效提高成功围捕的概率。针对其中的合围点协同部署问题，提出了一种基于有限时间一致性的合围点部署方法。首先，将合围点部署问题建模为单位圆上的覆盖控制问题，基于多智能体一致性理论提出的分布式控制协议能实现对单位圆的最优覆盖；通过引入时变可调增益策略，将闭环系统由渐进收敛调整为有限时间收敛，从而大大提高了合围点部署的效率。仿真结果表明，提出的分布式控制协议可有效实现多智能体合围点的快速自部署。