University of Chicago's "self-driving" lab automates experiments in quantum computing research
Researchers at the University of Chicago have built a "self-driving" lab that uses robotics and artificial intelligence to accelerate work that would take humans weeks or months to complete.
The autonomous lab at the University of Chicago Pritzker School of Molecular Engineering is the first "self-driving" lab of its kind in the Midwest.
Instead of getting someone or something from point A to point B like a delivery robot or autonomous car, the "self-driving" lab automates tedious labor in experiments that would take high-level researchers weeks or even months to complete.
"Basically combining artificial intelligence and robotic engineering," said Assistant Professor Shuolong Yang.
The prototype is a high-speed shortcut to the building blocks of future quantum computers and electronics. The system learns as it goes.
"I thought, why couldn't we use AI and robotics to automate this entire process?" said PhD student Bill Zheng. "The framework that we've developed here can help other material discovery as well."
It's not replacing the people who built it, just allowing them to work on more.
"They don't have to stay in the lab all day and all night doing this mechanical manual labor," Zheng said.
There are countless applications, but there's a good example playing out across the Chicago area of why this work is so important – data centers, with many concerned about the amount of energy and resources required to run them.
Experts estimate utility bills in the Chicago area could go up as much as $70 in the next three years because of data centers.
Yang said quantum materials they're building with the help of the "self-driving" lab at the University of Chicago could reduce the energy consumption needed for data centers.
"A superconductor mitigates that problem," Yang said.
The work at the "self-driving" lab is funded by a grant for another year, but will continue afterward with support from the university.