Distributed Nash Equilibrium Seeking in Merely Monotone Games Using an Event- and Switching-Activated Communication Scheme

• 成果类型: Article
• 署名作者: Ding, Lei; Ye, Maojiao; Han, Qing-Long; Jia, Guobiao
• 署名单位: Nanjing University of Posts & Telecommunications; Nanjing University of Science & Technology; Swinburne University of Technology
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2024.3433069
• 发表日期: 2025
• 页码: 471-478
• 关键词: games; switches; Nash equilibrium; TOPOLOGY; vectors; COSTS; Symmetric matrices; Distributed algorithm; Dynamic event-triggered control; leader-following consensus; monotone games; Nash equilibrium seeking

摘要

This article is concerned with distributed dynamic event-triggered Nash equilibrium seeking in merely monotone games under switching communication topologies. Based on a regularization algorithm, a gradient play and a leader-following consensus protocol, a distributed Nash equilibrium seeking strategy incorporating an event- and switching-activated communication scheme is established. More specifically, the regularization algorithm is involved to compensate for the absence of strict/strong monotonicity of game mapping and the gradient play is utilized to optimize players' objectives. The event-triggered leader-following consensus protocol is introduced to achieve distributed estimation of all players' actions, where communication transmissions among neighbors occur only when the preset event-triggering conditions are satisfied and communication topologies are switching. It is shown that the proposed seeking strategy can asymptotically steer players' actions to the least-norm Nash equilibrium of merely monotone games with reduced communication resources. An illustrative example on least-distance formation control for a network of unicycle-type mobile vehicles is given to validate the effectiveness of the proposed method.