AI-Assisted Design for Combinatorial Metamaterials

AI predicts the properties of metamaterials

Can you flatten a piece of 3D origami without damaging it? It is difficult to tell the answer just by looking at the origami design. Each fold must be compatible with the flattening process.

Here is a good example of a combinational problem. AMOLF and the UvA Institute of Physics have conducted a new research project that has shown how machine learning algorithms are able to answer such questions accurately and efficiently. This is expected to give a boost to the artificial intelligence-assisted design of complex and functional (meta)materials.

The team’s latest research, published this week in Physical Review Letters, tested how well AI can predict properties of metamaterials that combine mechanical and combinatorial elements.