Statistical Posture Prediction of Vehicle Occupants in Digital Human Modelling Tools

Abstract

When considering vehicle interior ergonomics in the automotive design and development process, it is important to be able to realistically predict the initial, more static, seated body postures of the vehicle occupants. This paper demonstrates how published statistical posture prediction models can be implemented into a digital human modelling (DHM) tool to evaluate and improve the overall posture prediction functionality in the tool. The posture prediction functionality uses two different posture prediction models in a sequence, in addition to the DHM tool´s functionality to optimize postures. The developed posture prediction functionality is demonstrated and visualized with a group of 30 digital human models, so called manikins, by using accurate car geometry in two different use case scenarios where the sizes of the adjustment ranges for the steering wheel and seat are altered. The results illustrate that it is possible to implement previously published posture prediction models in a DHM tool. The results also indicate that, depending on how the implemented functionality is used, different results will be obtained. Having access to a digital tool that can predict and visualize likely future vehicle occupants’ postures, for a family of manikins, enables designers and developers to consider and evaluate the human-product interaction and fit, in a consistent and transparent manner.