Scholarlink

<p>This paper tackles the challenge of predefined-time distributed optimization for nonlinear multi-agent systems with unmeasurable states. A novel predefined-time observer is designed using time-varying functions, thereby eliminating singularities inherent in conventional approaches. Unlike existing predefined-time results that address stability solely within finite intervals, a comprehensive full-period predefined-time observer-based control framework is developed. Furthermore, a generalized convexity condition is established for certain local functions, relaxing the traditional strong convexity requirement. The original optimization problem is first transformed into an unconstrained formulation using a quadratic penalty function. Subsequently, a distributed predefined-time optimization control scheme is synthesized via a back-stepping methodology. The proposed controllers ensure output consensus and achieve minimization of the global objective function within a predefined-time scheme. Rigorous stability analysis and numerical simulations demonstrate the efficacy of the developed methodology.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predefined-time observer-based distributed consensus optimization control of nonlinear multi-agent systems

  • 期刊论文

  • 发表日期：

    2025年8月29日

查看全文
  • Lina Guo,
  • Shuping He,
  • Jiasheng Shi,
  • Haijiao Yang

摘要

This paper tackles the challenge of predefined-time distributed optimization for nonlinear multi-agent systems with unmeasurable states. A novel predefined-time observer is designed using time-varying functions, thereby eliminating singularities inherent in conventional approaches. Unlike existing predefined-time results that address stability solely within finite intervals, a comprehensive full-period predefined-time observer-based control framework is developed. Furthermore, a generalized convexity condition is established for certain local functions, relaxing the traditional strong convexity requirement. The original optimization problem is first transformed into an unconstrained formulation using a quadratic penalty function. Subsequently, a distributed predefined-time optimization control scheme is synthesized via a back-stepping methodology. The proposed controllers ensure output consensus and achieve minimization of the global objective function within a predefined-time scheme. Rigorous stability analysis and numerical simulations demonstrate the efficacy of the developed methodology.