Scholarlink

This paper proposes a neural network-based direct iterative learning control (NN-DILC) method for unknown nonlinear multi-agent systems (MASs). First, the inherent consensus dynamics relationship is established between the consensus output and the control input. Then, an iterative linear data model is derived to equivalently reformulate the such a relationship. Further, a radial basis function neural network estimation algorithm along the iterative axis is designed to estimate unknown linearization parameters online, and subsequently, a NN-DILC scheme is proposed. The proposed NN-DILC is a purely data-driven control approach that does not rely on any model information. The simulation results verify the effectiveness of the proposed method by comparing with the model-free adaptive ILC.

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

Neural Network-Based Direct Iterative Learning Control for Multi-agent Systems

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Huiming Peng,
  • Na Lin

摘要

This paper proposes a neural network-based direct iterative learning control (NN-DILC) method for unknown nonlinear multi-agent systems (MASs). First, the inherent consensus dynamics relationship is established between the consensus output and the control input. Then, an iterative linear data model is derived to equivalently reformulate the such a relationship. Further, a radial basis function neural network estimation algorithm along the iterative axis is designed to estimate unknown linearization parameters online, and subsequently, a NN-DILC scheme is proposed. The proposed NN-DILC is a purely data-driven control approach that does not rely on any model information. The simulation results verify the effectiveness of the proposed method by comparing with the model-free adaptive ILC.