Deep Learning in Ergonomics: A 2024 Review on Advances, Applications, and Challenges

抄録

Deep learning (DL) is transforming ergonomics, offering new ways to assess and improve workplace safety and human performance. This review explores recent advancements in DL applications for ergonomics in 2024, focusing on cognitive workload assessment, posture recognition, human-robot collaboration, and real-time ergonomic monitoring. By analyzing recent studies from Google Scholar, we identify key trends, innovations, and challenges, including data limitations, model interpretability, and the need for human-centered AI design. Emerging technologies such as large multimodal models (LMMs) and digital human modeling (DHM) are also examined for their potential impact. This study provides a comprehensive overview of how DL is reshaping ergonomic assessments, offering valuable insights for researchers, industry professionals, and policymakers. Our findings highlight opportunities for further research and the practical implementation of AI-driven solutions to enhance workplace ergonomics and safety.