In June 2018, a small research team at the Wyss Institute of Harvard University made media headlines across the world as global aerospace firm Rolls-Royce publicized their joint research in building tiny robots capable of inspecting the insides of jet engines.
Our reporter Lindsay Clark caught up with former Wyss Institute research fellow Sébastien de Rivaz as he waited for publication of research based on the so-called SWARM robots in the American journal Science Robotics.
Lindsay Clark: How did you start working with small robots for the purpose of jet engine maintenance?
De Rivaz: The Wyss Institute already had a micro-robot platform called the Harvard Ambulatory Microrobot (HAMR) which has more than 60 joints including hip and leg joints that mimic those of cockroaches. A lot of research had been completed on the platform, although it was not able to ascend vertical walls or cling to surfaces upside-down. Rolls-Royce reached out to the team behind that to demonstrate the potential of collaborative “SWARM” robots that crawl through the insides of an engine, as part of its IntelligentEngine vision. I joined a few months after that and took charge of that particular project in about June 2017. The objective was to see if we could develop robots small enough to work together to inspect an engine for wear and damage. They would be delivered into the machine together using a special snake-like robot.
Lindsay Clark: What were the greatest challenges when it came to making the robots capable of inspecting jet engines?
De Rivaz: One of the biggest challenges is scaling. The robot we presented to the media, which appears in a few videos, is not the latest version. We had to reduce that scale by a factor of two, so it could fit through the side holes, about 10 mm. That reduction in scale makes them very difficult to manufacture, and we had to do some research to push that boundary. Another challenge is getting them to climb on vertical and inverted surfaces, but I can’t talk about that too much, as we are awaiting review of our paper in Science Robotics.
Lindsay Clark: Do you foresee mass adoption of these robots if they prove successful in the lab?
De Rivaz: At the moment it is kind of a niche product, more of a proof of concept in partnership with Rolls Royce. I’d say we are years away from mass adoption. Production at scale is another whole research topic. At the moment, the basic frame is manufactured in quite an unconventional way and after that we do everything by hand. Getting a higher yield from the whole assembly process, for mass production, that’s another question.
Lindsay Clark: Why do you mimic mechanism you find in nature?
De Rivaz: We were trying to reproduce aspects of what we find in nature. That is complex, but we also trust evolution. When most people see an insect, they want to squash it; we want to figure out the power sources and the actuation mechanism. Millions of years of evolution is a real inspiration. Trying to mimic that is the challenge.
De Rivaz left the SWARM project after the initial research phase and is now a product design engineer at Apple.