Skip to main navigation Skip to search Skip to main content

A Simulation Modeling Method for Electromechanical Control of Large-inertia Manipulator Based on Deep Koopman Operator

  • Zhuoqi Manthe
  • , Junyu Wu
  • , Xuanming Cao
  • , Yubin Liu*
  • *Corresponding author for this work
  • Harbin Institute of Technology
  • Harbin Engineering University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes a deep extended dynamic mode decomposition architecture with residual modules and feedback channels (RDEDMD) to construct a highdimensional linearized dynamic model of 6-DOF manipulator using Koopman operator. Basing on this foundation, the RKMPC model predictive control system is designed and integrated with the developed electromechanical simulation model for high-inertia manipulator, thereby establishing a unified electromechanical control simulation framework based on deep Koopman operators. Experimental results demonstrate that the proposed simulation model achieves significantly higher accuracy than traditional mechanismbased modeling approaches, while the RK-MPC controller exhibits superior prediction and control performance compared with conventional control methods, thus validating the effectiveness of the linearized model constructed by the RDEDMD.

Original languageEnglish
Title of host publicationProceedings - 2026 International Conference on Robotics, Automation and Intelligent Transportation Systems, RAITS 2026
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331558079
DOIs
StatePublished - 2026
Event2026 International Conference on Robotics, Automation and Intelligent Transportation Systems, RAITS 2026 - Xi'an, China
Duration: 23 Jan 202625 Jan 2026

Publication series

NameProceedings - 2026 International Conference on Robotics, Automation and Intelligent Transportation Systems, RAITS 2026

Conference

Conference2026 International Conference on Robotics, Automation and Intelligent Transportation Systems, RAITS 2026
Country/TerritoryChina
CityXi'an
Period23/01/2625/01/26

Keywords

  • 6-DOF Manipulator
  • Deep Learning
  • Electromechanical control simulation model
  • Koopman Operator

Fingerprint

Dive into the research topics of 'A Simulation Modeling Method for Electromechanical Control of Large-inertia Manipulator Based on Deep Koopman Operator'. Together they form a unique fingerprint.

Cite this