Related Experiment Video
Updated: Mar 12, 2026
07:25
An Objective and Child-friendly Assessment of Arm Function by Using a 3-D Sensor
Published on: February 12, 2018
7.3K
Validation of an ergonomic assessment method using Kinect data in real workplace conditions
Pierre Plantard1, Hubert P H Shum2, Anne-Sophie Le Pierres3
1FAURECIA Automotive Seating, ZI de Brières les Scellés, B.P. 89 91152, Etampes, France; M2S Lab., University Rennes 2, ENS Rennes, Avenue Robert Schuman, 35170, Bruz, France.
Applied Ergonomics
|November 9, 2016
Summary
This study enhances musculoskeletal disorder risk assessment using corrected Microsoft Kinect data for accurate worker posture analysis in cluttered environments. The method provides reliable Rapid Upper Limb Assessment (RULA) scores, improving ergonomic evaluations.
Area of Science:
- Ergonomics
- Occupational Health
- Biomechanics
Background:
- Assessing musculoskeletal disorder (MSD) risks in real workstations is challenging due to environmental clutter hindering accurate posture evaluation.
- Microsoft Kinect offers a marker-free, calibration-free solution but is susceptible to occlusions.
- Existing methods struggle with the dynamic and complex nature of real-world work environments.
Purpose of the Study:
- To evaluate the efficacy of a Rapid Upper Limb Assessment (RULA) ergonomic assessment using occlusion-resistant Kinect skeleton data correction in real work conditions.
- To determine if corrected Kinect data can accurately assess worker postures and provide reliable RULA scores in cluttered, non-laboratory settings.
- To explore the potential of Kinect technology for continuous, real-time MSD risk assessment.
Main Methods:
- Developed and applied an occlusion-resistant Kinect skeleton data correction method for posture estimation.
- Validated posture estimations against ground-truth data under standardized laboratory conditions.
- Compared RULA scores derived from corrected Kinect data with expert assessments in a cluttered workplace environment.
Main Results:
- The corrected Kinect data demonstrated improved accuracy in estimating worker postures compared to uncorrected data.
- RULA scores generated using the corrected Kinect data showed higher agreement with expert evaluations in real-world conditions.
- The system provided 30 Hz continuous posture information, suitable for both offline and real-time analysis.
Conclusions:
- The proposed method using corrected Kinect skeleton data significantly enhances the accuracy of RULA ergonomic assessments in challenging, real-world work environments.
- This technology offers a promising tool for ergonomists, enabling more objective and continuous musculoskeletal risk assessment.
- The findings open new perspectives for proactive MSD prevention through improved real-time and offline ergonomic analysis.
