Automatic detection system for real-time assessment of human work postures during hammering operation

Abstract

posture assessment. This paper presents the validation of a developed real-time ergonomic assessment knowledge-based system for use in the real-time evaluation of work postures on the shop floor and the provision of feedback to workers. The system is developed using a cost-effective, automatic-detection 3D motion sensor. The developed intelligent system utilizes the knowledge from health and safety guidelines, a set of rules and an inference engine, to automatically capture and assess workers’ postures and provide real-time feedback to the worker through an easy-to-understand user interface. Results of testing the developed system showed that the system can detect manual handling tasks such as hammering activity and give real-time feedback to the operator on the task detection, as confirmed by the ‘True’ value displayed for hammering, with detection confidences of approximately 0.4, 0.5 and 0.7. Results also showed that the system can assess work postures and provide real-time feedback to workers simultaneously with task detection. The system is beneficial to workers on the manufacturing shop floor as it can correct their work methods and alert them to adjust awkward postures that can lead to Work-Related Musculoskeletal Disorders.