A real-time posture assessment system based on motion capture data for manual maintenance and assembly processes

Dong Zhou, Chengzhang Chen, Ziyue Guo, Qidi Zhou, Dengwei Song, Aimin Hao

Research output: Contribution to journalArticlepeer-review

Abstract

As manufacturing processes of complex products become automated, manual operations still occupy a considerable portion of industrial maintenance and assembly (IMA), especially in the machinery and aerospace fields. Workers often exhibit awkward posture in IMA activities. In these scenarios, posture assessment is critical for improving the well-being of workers because awkward postures can lead to work-related musculoskeletal disorder (WMSD). Although there are several categorized WMSD risk assessment methods, limited evidence suggests that these methods are compliant for modern complex IMA scenarios. In this paper, a posture analysis system for manual operation (PASMO) is presented to monitor working postures and evaluate WMSD risks in IMA processes. The noninvasive depth sensor Kinect v2 and the rapid upper limb analysis (RULA) method are integrated to achieve this purpose. In the PASMO, the RULA is optimized and driven by motion capture (MoCap) data to make evaluating the WMSD risk of working postures more effective and accurate. Industrial and laboratory experiments are designed to verify the effectiveness and system performance of the PASMO. The results show that for the overall body and most joints, the scores obtained by the PASMO substantially agree with those obtained by the ground truth data (p < 0.01, κ = 0.65) under the real industrial environment. Because the experiments are conducted in real IMA scenarios with body occlusion, the results preliminarily prove the effectiveness of the PASMO for WMSD risk assessment in continuous IMA tasks.

Original languageEnglish
Pages (from-to)1397-1411
Number of pages15
JournalInternational Journal of Advanced Manufacturing Technology
Volume131
Issue number3-4
DOIs
StatePublished - Mar 2024
Externally publishedYes

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

Keywords

  • Industrial maintenance and assembly
  • Kinect
  • Musculoskeletal disorders
  • Working posture

Fingerprint

Dive into the research topics of 'A real-time posture assessment system based on motion capture data for manual maintenance and assembly processes'. Together they form a unique fingerprint.

Cite this