Fraunhofer Center for Assistive Information and Communication Solutions – AICOSDescription:
Manufacturing processes are based on human labour and the symbiosis between human operators and machines. The operators are required to follow predefined sequences of movements. The operations carried out at assembly lines are repetitive, being identified as a risk factor for the onset of musculoskeletal disorders.
Ergonomics plays a big role in preventing occupational diseases. Ergonomic risk scores measure the overall risk exposure of operators however these methods still present challenges: the scores are often associated to a given workstation, being agnostic to the variability among operators. Observation methods are most often employed yet require a significant amount of effort, preventing an accurate and continuous ergonomic evaluation to the entire population of operators. Finally, the risk’s results are rendered as index scores, hindering a more comprehensive interpretation by occupational physicians.
This dissertation developed a solution for automatic operator risk exposure in assembly lines. Three main contributions were presented: (1) an upper-body motion tracking algorithm which relies on inertial sensors to estimate the orientation of anatomical joints; (2) an adjusted ergonomic risk score; (3) an ergonomic risk explanation approach based on the analysis of the angular risk factors. Throughout the research, two experimental assessments were conducted: laboratory validation and field evaluation. The laboratory tests enabled the creation of a movements’ dataset and used an optical motion capture system as reference. The field evaluation dataset was acquired on an automotive assembly line and serve as the basis for an ergonomic risk evaluation study. The experimental results revealed that the proposed solution has the potential to be applied in a real environment. Through direct measures, the ergonomic feedback is fastened, and consequently, the evaluation can be extended to more operators, ultimately preventing, in long-term, work-related injuries.
Author: Sara Santos
Type: MSc thesis
Partner: Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa
Year: 2019