Consensus With a Linear Constraint

• 成果类型: Article
• 署名作者: Corless, Martin J.; Coduti, Leonardo
• 署名单位: Purdue University System; Purdue University
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2023.3315688
• 发表日期: 2024
• 页码: 645-650
• 关键词: Resource management; TOPOLOGY; Null space; Network topology; Behavioral sciences; Laplace equations; Consensus algorithm; consensus

摘要

In this article, we consider a network of agents running a linear or nonlinear consensus algorithm with the following goal: in addition to requiring that the outputs of all agents converge to the same value, we also require that a specified linear function of the agent states remains constant during the evolution of the consensus algorithm. To achieve this goal requires the construction of a matrix of weighting parameters with specific properties. In this article, we present a noniterative centralized algorithm and an iterative decentralized algorithm for determining the weighting parameters. In the decentralized algorithm, the weighting parameters are specified by the agents and each agent only specifies the weighting parameters associated with the agents to which it communicates. The results do not require that the communication graph of the network be bidirectional. The results of this article can also be applied to consensus problems where one wants to achieve consensus to a specified weighting of the initials states of the network.