ABSTRACT
This work investigates linear and non-linear parametric reduced order models (ROM) capable of replacing computationally expensive high-fidelity simulations of human body models (HBM) through a non-intrusive approach. Conventional crash simulation methods pose a computational barrier that restricts profound analyses such as uncertainty quantification, sensitivity analysis, or optimization studies. The non-intrusive framework couples dimensionality reduction techniques with machine learning-based surrogate models that yield a fast responding data-driven black-box model. A comparative study is made between linear and non-linear dimensionality reduction techniques. Both techniques report speed-ups of a few orders of magnitude with an accurate generalization of the design space. These accelerations make ROMs a valuable tool for engineers.
Subject(s)
Human Body , Machine Learning , Humans , UncertaintyABSTRACT
A new technique to accelerate the positioning of human body models (HBMs) by means of a dimensionality reduction of a database of precomputed simulations is presented. First, a set of important subspace deformation modes which are used to approximate the model's movements observed in the training simulations are computed. In the second step, a convex optimization problem is solved in order to obtain an optimal position of the human body model as described by the user. We apply the proposed method to a new reclined seating position of the Total Human Model for Safety (THUMS, v5).