二部图
多智能体系统
差别隐私
共识
计算机科学
协议(科学)
控制理论(社会学)
事件(粒子物理)
李雅普诺夫函数
控制(管理)
数学
理论计算机科学
算法
非线性系统
人工智能
量子力学
医学
物理
图形
病理
替代医学
作者
Jiayue Ma,Jiangping Hu
标识
DOI:10.1007/s40747-024-01430-2
摘要
Abstract This paper studies the problem of differentially private bipartite output consensus in continuous-time heterogeneous multi-agent systems (MASs) characterized by antagonistic interactions. A novel hybrid privacy-preserving event-triggered impulsive consensus protocol is introduced to protect the agent’s initial information from disclosure, which involves a discrete-time information transmission based on an event-triggering mechanism. Using stochastic Lyapunov method, sufficient conditions have been obtained to achieve mean square bipartite output consensus with a guaranteed level of privacy. Furthermore, the differential privacy of competitive agent pairs is exclusively secured by the proposed control scheme by injecting Laplace noise. The protocol also effectively prevents Zeno behavior by imposing a lower bound for impulsive intervals under all event-triggered conditions. A simulation example is provided to validate the effectiveness of the theoretical result.
科研通智能强力驱动
Strongly Powered by AbleSci AI