Neural-Network-Based Inverse Dynamic Solutions of an Asymmetric Parallel Robot

Qi Zou, Dan Zhang, Shuo Zhang, Yuancheng Shi, Guanyu Huang, Lijian Li

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

The dynamic models of parallel robots are more complicated due to the highly coupled kinematic chains compared with those of their counterparts-serial robot. The approximate dynamic models can be considered if the analytical solutions high cost in time due to the sophisticated calculations. To address this issue, this paper presents an alternative approach-the dynamic model based on the neural network. A three degrees-of-freedom parallel manipulator is introduced, followed by a detailed mathematical inverse dynamic model, which is replicated by three feedforward neural networks with various performance goals. A cascade forward neural network is also constructed to be compared with the feedforward neural network. The simulation results demonstrate the effectiveness of these proposed neural networks.

Original languageEnglish
Title of host publicationProceedings - 2024 3rd International Conference on Innovations and Development of Information Technologies and Robotics, IDITR 2024
EditorsDan Zhang, Hamid Reza Karimi
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages94-99
Number of pages6
ISBN (Electronic)9798350385694
DOIs
Publication statusPublished - May 2024
Event3rd International Conference on Innovations and Development of Information Technologies and Robotics, IDITR 2024 - Hong Kong, China
Duration: 23 May 202425 May 2024

Publication series

NameProceedings - 2024 3rd International Conference on Innovations and Development of Information Technologies and Robotics, IDITR 2024

Conference

Conference3rd International Conference on Innovations and Development of Information Technologies and Robotics, IDITR 2024
Country/TerritoryChina
CityHong Kong
Period23/05/2425/05/24

Keywords

  • cascade forward neural network
  • dynamic modelling
  • feedforward neural network
  • Parallel robot

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Information Systems
  • Control and Optimization
  • Modelling and Simulation

Fingerprint

Dive into the research topics of 'Neural-Network-Based Inverse Dynamic Solutions of an Asymmetric Parallel Robot'. Together they form a unique fingerprint.

Cite this