Automatic learning of climbing configuration space for digital human children model

Millions of children die from preventable injuries every year around the world. Environmental modification is one of the most effective ways to prevent these fatal injuries. The environment should be modified and products should be designed in ways that will reduce the risk of injury by taking child–environment and child–product interactions into account. However, it is still very difficult even for advanced simulation systems to predict how children interact with products in everyday life situations. In this study, we explored a data-driven method as a promising approach for simulating children’s interaction with products in everyday life situations. We conducted an observational study to collect data on children’s climbing behavior and developed a database on children’s climbing behavior to clarify a climbing configuration space, which enables the prediction and simulation of the possible climbing postures of children.