Type de document
Études primaires
Année de publication
2024
Langue
Anglais
Titre de la revue
Journal of Intelligent Manufacturing
Résumé
Disassembly acts as the initial phase in remanufacturing processes. Human–robot collaboration revolutionizes disassembly by enhancing efficiency and accuracy. The integration of industrial robotics—cobots included—for human safety is mandated by ISO 10218-2:2011. Furthermore, ISO/TS 15066:2016 requires conducting risk assessments to prevent accidents for cobotic systems. We integrate computer vision techniques, including YOLO V5 for human detection and color-based cobot arm identification, to monitor and assess hazards in real time. Laboratory experiments demonstrate the system’s effectiveness in identifying potential collisions and other safety risks, with a 99.9% accuracy rate for human detection. The system also exhibits fast detection times, with YOLO processing a single frame in 0.0089 s and color detection taking 0.0457 s. Additional tests under simulated dusty conditions showed minimal impact on detection accuracy. This research is relevant to stakeholders in manufacturing, safety engineering, and sustainability. By enhancing safety in human–robot collaborative disassembly, it addresses key issues in remanufacturing, recycling, and workplace safety, with broader implications for industrial safety. Graphical abstract: (Figure presented.) © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.
Mots-clés
Robot collaboratif, Collaborative robot, Évaluation du risque, Hazard evaluation, Gestion du risque, Risk management
Numéro de projet IRSST
n/a
Citation recommandée
Alenjareghi, M. J., Keivanpour, S., Chinniah, Y. A. et Jocelyn, S. (2024). Computer vision-enabled real-time job hazard analysis for safe human-robot collaboration in disassembly tasks. Journal of Intelligent Manufacturing. https://doi.org/10.1007/s10845-024-02519-8