Automated camera-based ergonomic evaluation of workplaces
Incorrect lifting and carrying of loads, e.g. at order picking workstations, can put a noticeable strain on the entire musculoskeletal system and the spine. Poor posture can lead to overloading, e.g. in the form of back problems. Small and medium-sized enterprises (SMEs) in particular cannot compensate for such illness-related absences with their company size, so that productivity and competitiveness suffer. In addition, expenses for convalescence and retraining must be made for sick employees, which also burden the economy as a whole. It is therefore in everyone’s interest that people are recognized as a valuable resource that should be treated with care. Previous ergonomic evaluation methods (EBM) are always based on the subjective categorization of movements by the evaluating employee. Therefore, on the one hand, a method should be found that enables SMEs to carry out cost-effective ergonomics assessments. On the other hand, this method must offer a repeatable ergonomic evaluation that is objective and independent of the person carrying out the test. This is to be achieved using a universal, camera-based, real-time capable EBM. In addition to a direct and localized estimation of the stress on the human skeleton thanks to machine learning methods, direct feedback to the employee should be made possible in the event of excessive stress or incorrect posture. This should support the learning of a health-friendly execution of the work processes.
Federal Ministry of Economics and Climate Protection (BMWK), Program for the funding of collaborative industrial research (IGF)
- Bundesvereinigung Logistik (BVL) e.V. (project coordinator)
- IPH – Institut für Integrierte Produktion Hannover gGmbH
Prof. Dr.-Ing. Bodo Rosenhahn