The Rapid Prototyping Laboratory (RPL) at the U.S. Department of Energy’s Argonne National Laboratory is a bright, high-ceilinged room alive with the whir of robotic arms, the hum of 3D printers, and the gust of cooling fans. Young researchers huddle around laptops and workbenches. They program robots, test cameras and sensors, build virtual-reality models of workspaces and print new mechanical attachments.
These college interns, master’s students and Ph.D. candidates are laying the groundwork not only for their own future careers in laboratory science, but for the future of science itself. That’s because the RPL is at the heart of an Argonne-wide effort called autonomous discovery, which incorporates robotics, machine learning and artificial intelligence. Autonomous discovery will speed the rate of scientific discovery while freeing up scientists’ hands and brains to work on tasks that only humans can accomplish.
“Maybe it’s the right solution the fifth time. That’s what the RPL is all about. Figuring out that solution.” — Casey Stone, Argonne computational scientist
Interns collaborating, tinkering and prototyping in the RPL might design and print a custom attachment so a robotic arm can pick up a delicate petrie dish without smashing it, or test cameras that help robots see and avoid obstacles in a lab.
Rafael Vescovi calls these steps “microtasks.” Vescovi is a data scientist with the Data Science and Learning division. He and computational scientist Casey Stone are responsible for helping interns learn and grow during their time at Argonne. Vescovi and Stone work on ways to implement autonomous systems alongside existing instruments in Argonne’s labs and experiments. Vescovi says that chaining micro-projects together makes the long-term vision of autonomous discovery possible.
He acknowledges that it can be challenging for interns to parachute into such a big project. “But working in a group and working on a scientific or on a research and development environment,” he said, “I think they get a lot from that.”
Vescovi should know. He joined Argonne as an undergraduate physics student conducting research at the Advanced Photon Source in 2010 and started automating aspects of his experiments to improve efficiency.
Along with teaching students practical aspects of being a working researcher — like polishing their CVs, turning in weekly reports on time and posting their portfolios online — Vescovi and Stone also encourage students to follow their passion.
The student researchers come from a variety of backgrounds, from biology to computer science to esports. They’ve been inspired by everything from great teachers to video games to TV shows like Myth Busters. And their interests show in their research.
Kendrick Xie is a junior at the University of Chicago majoring in computer science. He’s also a competitive gamer. One of his first projects was to program a LoCoBot so he could control it with a PlayStation 4 controller.
Xie notes that if you break something in real life, you have to figure out how to fix it. “There are no safeguards to keep you from crashing or accidentally jamming the arm into the floor.” His real-life crash test of the robot caused a crack in the robot’s gripper, so he came to the RPL to 3D print a new one.
Xie, like many of the RPL interns, was new to 3D printing. But breaking things and learning new skills to fix them are part of the learning process.
Rory Butler says that’s part of the fun of working at the RPL. Butler is a computer science Ph.D. candidate doing master’s research in applied robotics at UChicago.
Butler was always interested in robotics. Now he uses virtual reality to build laboratory environments where his teammates test robot functions.
“Having this group of people with a very diverse set of experiences and education allows us to work with each other and accomplish things more quickly,” said Butler. “And it’s also a whole lot of fun because if there’s no set way of doing things, then you’re free to explore and try just wild and crazy ideas and see what sticks.”
In pivotal discovery, there is no roadmap
Casey Stone oversees the RPL and works with Vescovi mentoring the interns. With a background in biological sciences, Stone is familiar with the precise and methodical aspects of laboratory research. As an undergraduate, she majored in biology but realized she didn’t enjoy bench science, especially repetitive tasks like pipetting: using a small suction device to transfer liquid samples during experiments. This hand cramping, repetitive motion can occupy a biologist for several hours each day.
“It’s just not my idea of fun,” Stone said. “I went back to school for computer science, so that I could also do biology, but maybe not hands-on all day. That’s how I got into automating biology.”
Stone loves the enthusiasm and teamwork she sees when the RPL student researchers attack a challenge. She points out that they rarely find the solution or create the perfect prototype on the first try. “Maybe it’s the right solution the fifth time. That’s what the RPL is all about. Figuring out that solution.”
Stone laughs as she notes that that process is like a lot of peoples’ career paths in science. “Try one thing. It doesn’t work. Try something else.”
Although the students’ projects are paving the way for larger discoveries with autonomous discovery, they do have “Eureka” moments of their own. Over the summer, Gillian Camacho, Arleen Hidalgo and Halona Dantes worked together to test the findings of a paper on automating experiments for growing E. coli in different nutrient environments.
Camacho is a junior at Penn State studying chemistry and forensics. Hidalgo is a senior at the University of Puerto Rico at Mayaguez studying industrial microbiology. Dantes is a senior at the University of North Carolina at Chapel Hill studying physics. Before coming to Argonne, none of them had significant experience with robotics or programming, but they loved the challenge of stretching outside of their focus areas.
“We all had different areas that we were stronger in. And I think that helped a lot,” said Camacho. “We ended up moving pretty quickly once we learned how to communicate with each other.”
The women were thrilled the first time they saw their robot run their commands to successfully complete a task. That thrill came not just from witnessing their first successful run, but also from knowing that what they were doing would help researchers achieve a lab-wide goal.
“It’s exciting and it’s something new and we’re contributing for it to get better,” said Hidalgo.
Dantes immediately saw autonomous discovery’s benefits to fellow researchers outside of the physics labs she’s used to. “I’ve learned how to pipette this summer,” she said. “And it’s very tedious. So many things can go wrong. I’m excited that people won’t have to do that in the future!”
“That’s so cool!”
Rafael Vescovi says that the investment Argonne makes in interns pays off. He has watched several interns grow from beginners to specialists. “Once they are trained, there’s always the chance that one of them will stay.”
Several young researchers said Argonne’s friendly and open culture and its world-class facilities had them considering long-term careers at Argonne.
Kyle Hippe is a first-year research scientist at Argonne who recently completed an internship at the lab. Hippe is interested in exploring the frontiers of artificial intelligence and its applications to biology. He came to the lab because of one thing: supercomputers.
Hippe says it has been thrilling to work on machines like the Polaris supercomputer. “This is what I dreamed of when I was a kid and that’s so cool!” But he was surprised to find that Argonne offered him the flexibility to explore different aspects of science.
“I’m not a roboticist, but I’ve gotten an opportunity to work in the RPL and touch some of these things that may be super useful for my future. The wealth of opportunities is just incredible at Argonne.”
Argonne National Laboratory seeks solutions to pressing national problems in science and technology. The nation’s first national laboratory, Argonne conducts leading-edge basic and applied scientific research in virtually every scientific discipline. Argonne researchers work closely with researchers from hundreds of companies, universities, and federal, state and municipal agencies to help them solve their specific problems, advance America’s scientific leadership and prepare the nation for a better future. With employees from more than 60 nations, Argonne is managed by UChicago Argonne, LLC for the U.S. Department of Energy’s Office of Science.
The U.S. Department of Energy’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.