ABSTRACT

When evaluating human-machine interaction in a virtual environment using Digital human modelling (DHM) it is important to ensure that the predicted motions lie within the range of behavioural diversity for different people within a population. This paper presents a study in which a comparison is made between motions predicted by the DHM tool IMMA (Intelligently Moving Manikin) and motions from real humans stored in a motion database. Results show similar motions but the predicted motions were in total statistically significantly different compared to the motions performed by real persons. The differences are most likely due to the balance function and joint constraints that the IMMA tool uses for predicting motions. Differences can also be due to other factors, aside of body size, such as age, gender or strength that affects the movement behaviour.