Multiobjective Distributed Optimization via a Predefined-Time Multiagent Approach

• 成果类型: Article
• 署名作者: Liu, Yang; Xia, Zicong; Gui, Weihua
• 署名单位: Zhejiang Normal University; Zhejiang Normal University; Central South University
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2023.3244122
• 发表日期: 2023
• 页码: 6998-7005
• 关键词: Distributed optimization; Multiagent system (MAS); multiobjective optimization; predefined-time optimization; time-base generator (TBG)

摘要

In this study, we propose a predefined-time multi-agent approach for multiobjective optimization. Predefined-time optimization is an optimization approach that can converge to a state that is extremely close to an optimal solution at a given time. A time-base generator is derived and applied to the optimization approaches for achieving predefined-time optimization. The multi-objective optimization problem is reformulated as a distributed optimization problem and, thus, solved in a private and safe manner. For distributed optimization, a multiagent system with time-base generators is developed for predefined-time optimization, and its convergence and speed are proven. Several examples confirm the validity of the results.