BME Professor Receives NSF Funding to Collaborate on National Network of AI-Programmable Cloud Laboratories
The National Science Foundation (NSF) is investing as much as $400 million to create a nationwide network of AI-programmable cloud laboratories. Adam Gormley, an associate professor in the Department of Biomedical Engineering, is a co-PI on COAST Programmable Cloud Laboratories (PCL)—a four-year $20 million NSF-funded project led by the University of California, Santa Barbara (USCB), combining robotics, AI, and advanced technologies to enable researchers to conduct experiments from anywhere.
Gormley is receiving $4.5 million as a co-PI to support Rutgers' Scientific Hub for Orchestrated Research and Experimentation, or SHORE. A companion development site to COAST PCL, SHORE will be a testbed for software, robotic procedures, instrument connections, and experimental workflows before they are introduced to COAST PCL at UCSB.
"We are proud to play such a critical role in the development of the NSF's transformative network of cloud laboratories," says SoE Dean Alberto Cuitiňo. "Researchers across the nation will have access to individually tailored programmable AI to meet their objectives at every stage—from project design to project evaluation—of their work."
Gormley reports that his group has already developed three connected laboratory software programs. Geppetto will translate a scientist's goal into an experimental protocol. PolyCraft will coordinate instruments and robotics, while Pinocchio will analyze the results, compare them with original hypotheses, and suggest follow-up experiments.
"Once workflows and systems have been developed here at SHORE, we can then push those updates to the facility in California to avoid downtime with the user facility at UCSB," says Gormley.
Most importantly, this arrangement gives the team a place to test new capabilities without interrupting projects underway for COAST PCL users.
A Revolutionary Research Connection
With the help of SHORE, scientists searching for stronger composite materials, more durable medical polymers, and more would be able to test hundreds of possibilities at UCSB without leaving their home institutions—and dramatically transform their research experiences.
UCSB is leading the Computation-Optimized Automated Soft materials Technology Programmable Cloud Laboratory, or COAST PCL—one of twenty remotely accessible and interconnected NSF-funded research nodes to create a national resource for automated polymer and soft-materials research.
It will ultimately connect instruments, robotics, software, data systems, and artificial intelligence in a cloud-accessible laboratory. Researchers will be able to design experiments remotely, have automated systems carry them out, and leverage AI to decide what should be tested next.
COAST PCL will also preserve useful information, including the unsuccessful formulations, routine trials, processing conditions, and incremental adjustments that rarely appear in published papers.
By recording all data associated with each sample, COAST PCL will create a body of knowledge that researchers and the AI systems assisting them can draw upon over time.
A Faster Way to Search for New Materials
Polymers are long chains of molecular building blocks that form materials used in medicine, electronics, energy systems, aerospace, construction, and consumer products.
Researchers regularly alter polymers to determine whether a polymer is strong of flexible, heat resistant, compatible with biological tissue, able to conduct ions, or responsive to its surroundings. Currently, polymer discovery is a primarily lengthy manual operation that produces results that don't readily compare with those produced by other labs.
COAST PCL would make that search faster, broader, and more consistent. A researcher could remotely design an experiment, and the automated COAST PCL and other PCL systems could do the rest quickly and efficiently.
A National Resource for Materials Discovery
Remote access will extend COAST PCL's reach to researchers unable to travel to UCSB or afford a comparable laboratory infrastructure. Scientists at undergraduate-focused institutions, emerging research universities, national laboratories, and industry could thus use equipment and expertise that would otherwise be out of geographical or financial reach.
The broader NSF initiative is intended to support a connected national system, not a collection of independent facilities. Participating nodes, or hubs, will work toward developing compatible data formats, software, laboratory procedures, and methods for determining whether findings can be reproduced elsewhere.
According to Gormley, helping to establish those practices brings significant responsibility.
"I'm incredibly humbled because, if we do this right, we have the opportunity to lay the foundation for what may eventually be a distributed infrastructure of small and large self-driving cloud laboratories," he says. "As well as an infrastructure that could continue to expand the discovery and impact of new materials for decades to come."
