Skip to main navigation Skip to search Skip to main content

Research on Human-Machine Task Collaboration Based on Action Recognition

  • Jihong Yan
  • , Shenyi Yan
  • , Lizhong Zhao
  • , Zipeng Wang
  • , Yun Liang
  • School of Mechatronics Engineering, Harbin Institute of Technology

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

Abstract

A human-machine collaboration method based on real time action recognition is presented to achieve the collaboration between automated guided vehicles (AGVs) and human operators through the perception of tasks conducted by human operators. The process of action recognition is accomplished by a Kinect vision sensor and convolutional neural networks (CNN), and the methods of multiple features fusion and hierarchical cluster are applied in the process of action recognition. The utilization of action recognition technology can effectively shorten the expected waiting time, and can avoid the assembly line stagnation caused by human errors to a great extent. Finally, the experiment result indicates that the overall efficiency has been increased significantly after applying the method.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE International Conference on Smart Manufacturing, Industrial and Logistics Engineering, SMILE 2019
EditorsChen-Fu Chien, Renzhong Tang, Runliang Dou, Jei-Zheng Wu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages117-121
Number of pages5
ISBN (Electronic)9781538679982
DOIs
StatePublished - Apr 2019
Externally publishedYes
Event2019 IEEE International Conference on Smart Manufacturing, Industrial and Logistics Engineering, SMILE 2019 - Hangzhou, China
Duration: 20 Apr 201921 Apr 2019

Publication series

NameProceedings - 2019 IEEE International Conference on Smart Manufacturing, Industrial and Logistics Engineering, SMILE 2019

Conference

Conference2019 IEEE International Conference on Smart Manufacturing, Industrial and Logistics Engineering, SMILE 2019
Country/TerritoryChina
CityHangzhou
Period20/04/1921/04/19

Keywords

  • action recognition
  • human error
  • human-machine collaboration
  • hybrid assembly line
  • time allowance

Fingerprint

Dive into the research topics of 'Research on Human-Machine Task Collaboration Based on Action Recognition'. Together they form a unique fingerprint.

Cite this