Inverse Design of Inflatable Soft Membranes Through Machine Learning

Across fields of science, researchers have increasingly focused on designing soft devices that can shape-morph to achieve functionality. However, identifying a rest shape leads target 3D upon actuation is non-trivial task involves inverse design capabilities. In this study, simple and efficient platform presented pre-programmed shapes starting from 2D planar composite membranes. By training neural networks with small set finite element simulations, the authors are able obtain both optimal for pixelated elastomeric membrane inflation pressure required it morph into shape. The proposed method has potential be employed at multiple scales different applications. As an example, shown how these inversely designed membranes used mechanotherapy applications, by stimulating certain areas while avoiding prescribed locations.