Distributed Fixed-time Converging Nash Equilibrium Seeking for Non-cooperative Games over Multi-agent Networks and Applications

Project: Research

View graph of relations

Description

Distributed Nash equilibrium seeking in games over multi-agent networks has recently attracted considerable interest in research communities due to its distinctive advantages such as scalability, robustness, adaptability, and privacy protection, and also due to its great potential in wide engineering applications such as power allocation among microgrids, energy consumption games of heating-ventilation-air-conditioning (HVAC) systems and plug-in-hybrid-electric-vehicle (PHEV) systems, and formation control of unmanned aerial or surface vehicles (UAVs or USVs). Many effective and efficient approaches have been developed. Most of existing works focus on distributed Nash equilibrium seeking with asymptotical or exponential convergence over undirected network topologies though some of them indeed consider finite-time convergence or directed network topologies. Those existing works with asymptotical or exponential convergence only ensure that Nash equilibriums are reached when times goes to infinity. For those approaches with finite-time convergence, the settling time is nevertheless dependent on their initial states and can grow unbounded with the increasing distance of the initial states from equilibrium points. In addition, most of existing works concentrate on distributed Nash equilibrium seeking under ideal network scenarios though some of them do consider simple non-ideal scenarios such as network-induced uniform transmission delays. However, many practical applications call for fixed-time convergence instead of finite-time, exponential or asymptotical convergence, that is, Nash equilibrium seeking is achieved within a pre-specified time which is independent of initial conditions and implies better transient performance. Moreover, network-induced transmission delays are often non-uniform, time-varying, or even unbounded, and they are known to have significant adverse effects on the concerned systems, even worse, to disrupt their intended functions or services. Motivated by these observations, this project proposes to investigate and develop novel approaches and methodologies for distributed Nash equilibrium seeking with fixed-time convergence in non-cooperative games over multi-agent networks under undirected/directed network topologies, and/or network-induced transmission delays. The outcomes of this project are expected to enrich theoretical foundation on distributed Nash equilibrium seeking in noncooperative games over multi-agent networks, and, in turn, the proposed approaches and methodologies will provide more design tools and guidelines for distributed fixed-time converging Nash equilibrium seeking for non-cooperative games over multi-agent networks in practical engineering applications. 

Detail(s)

Project number9043529
Grant typeGRF
StatusActive
Effective start/end date1/01/24 → …