Skip to main navigation Skip to search Skip to main content

Neural control for constrained human-robot interaction with human motion intention estimation and impedance learning

  • University of Science and Technology Beijing
  • University of Sussex
  • South China University of Technology
  • Southeast University, Nanjing

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

4 Scopus citations

Abstract

In this paper, an impedance control strategy is proposed for a rigid robot collaborating with human by considering impedance learning and human motion intention estimation. The least square method is used in human impedance identification, and the robot can adjust its impedance parameters according to human impedance model for guaranteeing compliant collaboration. Neural networks (NNs) are employed in human motion intention estimation, so that the robot follows the human actively and human partner costs less control effort. On the other hand, the full-state constraints are considered for operational safety in human-robot interactive processes. Neural control is presented in the control strategy to deal with the dynamic uncertainties and improve the system robustness. Simulation results are carried out to show the effectiveness of the proposed control design.

Original languageEnglish
Title of host publicationProceedings - 2017 Chinese Automation Congress, CAC 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2682-2687
Number of pages6
ISBN (Electronic)9781538635247
DOIs
StatePublished - 29 Dec 2017
Externally publishedYes
Event2017 Chinese Automation Congress, CAC 2017 - Jinan, China
Duration: 20 Oct 201722 Oct 2017

Publication series

NameProceedings - 2017 Chinese Automation Congress, CAC 2017
Volume2017-January

Conference

Conference2017 Chinese Automation Congress, CAC 2017
Country/TerritoryChina
CityJinan
Period20/10/1722/10/17

Keywords

  • adaptive control
  • full-state constraints
  • human-robot interaction (HRI)
  • impedance learning
  • motion intention estimation
  • neural networks (NNs)

Fingerprint

Dive into the research topics of 'Neural control for constrained human-robot interaction with human motion intention estimation and impedance learning'. Together they form a unique fingerprint.

Cite this