Support Vector Regression and Robust Right Coprime Factorization Based Nonlinear Internal Model Control Design for Soft Robotic Finger

カテゴリ：研究会(論文単位)

論文No：CT25120

グループ名：【C】電子・情報・システム部門 制御研究会

発行日：2025/12/24

タイトル(英語)：Support Vector Regression and Robust Right Coprime Factorization Based Nonlinear Internal Model Control Design for Soft Robotic Finger

著者名：An Zizhen(東京農工大学),鄧 明聡(東京農工大学)

著者名(英語)： Zizhen An(Tokyo University of Agriculture and Technology),Mingcong Deng(Tokyo University of Agriculture and Technology)

キーワード：ソフトロボットアクチュエータ,オペレータ,ＳＶＲ,非線形制御,ロバスト既約分解,ロバスト制御,Soft robotic finger,operator,Support Vector Regression,nonlinear control,robust coprime factorization,robust control

要約(日本語)：A new control design is proposed in this paper to address the practical implementation of SVR based learning plants considering both stability and tracking performance. Specifically, in this design framework, robust right coprime factorization is selected to stabilize the learning plants from a perspective of input/output mapping rather than mathematical formulation. Additionally, while the stabilized learning plant is explored by RRCF, desired tracking performance is realized by an operator based nonlinear internal model control (IMC) design. Eventually, practical application on a soft robotic finger system is conducted, which indicates the better performance of using the controlled SVR based learning plant and the feasibility of the proposed framework.

要約(英語)：A new control design is proposed in this paper to address the practical implementation of SVR based learning plants considering both stability and tracking performance. Specifically, in this design framework, robust right coprime factorization is selected to stabilize the learning plants from a perspective of input/output mapping rather than mathematical formulation. Additionally, while the stabilized learning plant is explored by RRCF, desired tracking performance is realized by an operator based nonlinear internal model control (IMC) design. Eventually, practical application on a soft robotic finger system is conducted, which indicates the better performance of using the controlled SVR based learning plant and the feasibility of the proposed framework.

本誌：2025年12月27日制御研究会

本誌掲載ページ：37-42p

原稿種別：英語

PDFファイルサイズ：1,580Kバイト