Min-Max Consensus Algorithm for Multi-agent Systems Subject to Privacy-Preserving Problem

• Published in: Neural Information Processing Vol. 11307; pp. 132 - 142
• Main Authors: Wang, Aijuan, Mu, Nankun, Liao, Xiaofeng
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2018; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Min-max consensus algorithm; Multi-agent systems; Privacy-preserving
• ISBN: 9783030042387; 3030042383
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-04239-4_12

This paper proposes a privacy-preserving min-max consensus algorithm for discrete-time multi-agent systems, where all agents not only can reach a common state asymptotically, but also can preserve the privacy of their states at each iteration. Based on the proposed algorithm, the detailed consensus analysis is developed, including the impossibility of finite time convergence and the sufficient condition of consensus. Moreover, the privacy-preserving analysis is provided to guarantee the reliability of our privacy-preserving scheme. Finally, a numerical simulation is performed to demonstrate the correctness of our results.