Skip to main navigation Skip to search Skip to main content

Neural Network-Based Terminal Sliding Mode Control for Insulator Cleaning Robots

  • School of Astronautics, Harbin Institute of Technology
  • Jianghuai Advance Technology Center

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

Abstract

Insulator cleaning robot operates in challenging and unpredictable conditions, whereas the conventional control algorithms are not easy to adapt to the complex disturbances. To deal with this issue, in this paper, a two-degree-of-freedom Lagrangian dynamics model is derived and a radial basis function (RBF) neural network based non-singular fast terminal sliding mode (NFTSM) controller is designed for the insulator cleaning robot. The NFTSM is designed for tracking the robot’s reference trajectory, and RBF neural network is introduced to compensate external disturbances and internal unmodelled dynamic, which further improves both the dynamic and steady state performance of the robot. Finally, simulation results demonstrate the advantages of the proposed controller in insulator cleaning robot trajectory tracking control.

Original languageEnglish
Title of host publicationProceedings - 2024 IEEE 22nd International Conference on Industrial Informatics, INDIN 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331527471
DOIs
StatePublished - 2024
Externally publishedYes
Event22nd IEEE International Conference on Industrial Informatics, INDIN 2024 - Beijing, China
Duration: 18 Aug 202420 Aug 2024

Publication series

NameIEEE International Conference on Industrial Informatics (INDIN)
ISSN (Print)1935-4576

Conference

Conference22nd IEEE International Conference on Industrial Informatics, INDIN 2024
Country/TerritoryChina
CityBeijing
Period18/08/2420/08/24

Keywords

  • Sliding mode control
  • neural network
  • robotic systems

Fingerprint

Dive into the research topics of 'Neural Network-Based Terminal Sliding Mode Control for Insulator Cleaning Robots'. Together they form a unique fingerprint.

Cite this