Specified-Time Distributed Nash Equilibrium Seeking in Aggregative Games Over Directed Communication Networks

• 成果类型: Article
• 署名作者: Tao, Qianle; Liu, Yongfang; Zhao, Yu; Xian, Chengxin; Wen, Guanghui; Chen, Guanrong
• 署名单位: Northwestern Polytechnical University; City University of Hong Kong; Southeast University - China; City University of Hong Kong
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2026.3662268
• 发表日期: 2026
• 关键词: multiagent systems; consensus; algorithms

摘要

This article addresses specified-time distributed Nash equilibrium seeking (ST-DNES) problems for aggregative games over unbalanced directed communication networks. Each player's cost function is influenced by both its own strategy and an aggregate of all other players' strategies. First, a new research framework is developed, named specified-time collaborative planning and seeking. Under this framework, a class of ST-DNES algorithms is designed with a specified-time distributed average estimator (ST-DAE) over undirected networks based on Pontryagin's maximum principle. Then, to deal with the asymmetry of unbalanced directed networks, a couple of improved ST-DNES algorithms are developed with two different types of ST-DAEs, called integral-surplus and balance-compensator-based ST-DNES algorithms, respectively. With the help of these two ST-DAEs, the ST-DNES problem over unbalanced directed networks is successfully solved. It is noticed that, the balance-compensator-based ST-DNES algorithm needs to sort all players with different numbers if allowed. The integral-surplus-based ST-DNES algorithm does not require sorting but needs out-degree information of players in the network. One may select different algorithms in different situations. Finally, the proposed ST-DNES algorithms are used to solve an energy consumption game problem, which proves their effectiveness. It is highlighted that this is the first time to establish the specified-time convergence of aggregative games over unbalanced directed communication networks.