A human-like dynamic Digital Human Model controller for ergonomic assessment of workstations

Digital humans are software tools used for work equipment design. They can simulate the future workers’ activity and assess the workplace ergonomics. They therefore offer genuine advantages when prevention (safety and health) is to be integrated as of the earliest phases of design. However they are complex to use and, in certain cases, may lead to underestimation of the constraints and exertions of the future operators.
This article presents the development of dynamic control algorithms and models that are inspired by human behaviour, in particular by the human capacities for anticipation, retroaction, and learning. They make it possible to animate a digital human automatically and as realistically as possible, in order to make a reliable ergonomic assessment. These developments have been validated by means of biomechanical data from experimental situations. This work will be supplemented in order to take into account movement variability of the future operators as of the earliest phases of design for a work situation.