Inverse Design of Inflatable Soft Membranes Through Machine Learning

Antonio Elia Forte*, Paul Z. Hanakata, Lishuai Jin, Emilia Zari, Ahmad Zareei, Matheus C. Fernandes, Laura Sumner, Jonathan Alvarez, Katia Bertoldi

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

26 Citations (Scopus)
62 Downloads (Pure)


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

Original languageEnglish
Article number2111610
Number of pages1
JournalAdvanced Functional Materials
Issue number16
Early online date10 Jan 2022
Publication statusPublished - 19 Apr 2022


Dive into the research topics of 'Inverse Design of Inflatable Soft Membranes Through Machine Learning'. Together they form a unique fingerprint.

Cite this