Skip to main navigation Skip to search Skip to main content

ROBOTIC ARM TRAJECTORY TRACKING CONTROL BASED ON RBF NEURAL NETWORK ESTIMATION OF DYNAMIC PARAMETERS

  • Nannan Du
  • , Liang Yan*
  • , Tiantian Wang
  • , Suwan Bu
  • , Chris Gerada
  • , Xiaoshuai Liu
  • , Xuxu Yang
  • , Haien Li
  • *Corresponding author for this work
  • Beihang University
  • University of Nottingham

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

Abstract

In robotic arm operations, after planning the desired motion trajectory, accurate and stable trajectory tracking control is essential. However, in actual control, there are issues such as the uncertainty of the dynamic parameters and susceptibility to external disturbances during motion. Addressing these issues, this paper employs RBF neural networks to estimate dynamic parameters and utilizes an adaptive controller to achieve online trajectory tracking. The dynamic approximation capability and adaptability of this method enhance the real-time performance and disturbance rejection of robotic arm trajectory tracking control. Finally, the availability of this method is verified by experiments in the simulation environment.

Original languageEnglish
Title of host publicationCSAA/IET International Conference on Aircraft Utility Systems, AUS 2024
PublisherInstitution of Engineering and Technology
Pages260-264
Number of pages5
Volume2024
Edition13
ISBN (Electronic)9781837242108
DOIs
StatePublished - 2024
Event2024 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2024 - Xi�an, China
Duration: 16 Aug 202419 Aug 2024

Conference

Conference2024 CSAA/IET International Conference on Aircraft Utility Systems, AUS 2024
Country/TerritoryChina
CityXi�an
Period16/08/2419/08/24

Keywords

  • Dynamic parameters estimation
  • RBF neural network
  • robotic arm
  • trajectory tracking control

Fingerprint

Dive into the research topics of 'ROBOTIC ARM TRAJECTORY TRACKING CONTROL BASED ON RBF NEURAL NETWORK ESTIMATION OF DYNAMIC PARAMETERS'. Together they form a unique fingerprint.

Cite this