"AI gives you a treasure map, and the scientist needs to find the treasure," says Miguel Bessa, one of the researchers behind a new AI-created super-compressible material, and the lead author of a paper on the topic.
Bessa, alongside a team of researchers from TU Delft, has developed a new super-compressible yet robust material without carrying out any experimental tests at all. All they used was artificial intelligence (AI).
An AI-made material
As the researchers' paper says, "designing future‐proof materials goes beyond a quest for the best."
"The next generation of materials needs to be adaptive, multipurpose, and tunable. This is not possible by following the traditional experimentally guided trial‐and‐error process, as this limits the search for untapped regions of the solution space."
The solution to this? Artificial intelligence, the researchers say.
What the scientists did was use "a computational data‐driven approach" to explore the feasibility of a new meta-material concept.
By using AI, they could adapt the concept material to different target properties, choice of base materials, length scales, and manufacturing processes.
Inspired by solar sails
The work was inspired by Bessa's time at the California Institute of Technology. While there, he noticed a satellite structure at the Space Structures Lab, that was able to open out, large, expansive solar sails from within a very small storage space.
Bessa wondered if this type of highly compressible design could be compressed into an even smaller space. "If this was possible, everyday objects such as bicycles, dinner tables, and umbrellas could be folded into your pocket," he said in a press release.
He wondered if it would be possible to design a highly compressible, yet strong material that could be compressed into a small fraction of its volume. "If this was possible, everyday objects such as bicycles, dinner tables, and umbrellas could be folded into your pocket."
Reducing the need for experimentation
However, "metamaterial design has relied on extensive experimentation and a trial-and-error approach," Bessa says. "We argue in favor of inverting the process by using machine learning for exploring new design possibilities while reducing experimentation to an absolute minimum."
Using machine learning, Bessa and his team fabricated two designs of different sizes that transform brittle polymers into lightweight, recoverable materials that are super-compressible. One design was built for strength and the other for maximum compressibility.
Yet, Bressa argues that the real achievement in the team's work is in the method of creation, not the material itself. As he puts it, "data-driven science will revolutionize the way we reach new discoveries, and I can't wait to see what the future will bring us."