A team led by MIT engineers has built a robot that does something most diving birds cannot: launch itself directly out of the water and into flight without paddling first. The finding, published in Science on July 9, 2026, is a small mechanical breakthrough wrapped around a genuinely notable research-administration story — the paper’s co-authors span MIT, a Swiss engineering powerhouse, and a small tribal college in Washington State.
No paddling required
The robot, called the FAAV (flapping-wing aerial-aquatic vehicle), weighs less than 300 grams — about half a pound — and is built around a central fuselage, two flexible flapping wings coated with hydrophobic nanoparticles, and a motorized steerable tail. Researchers tested three wingspans (60, 80, and 100 centimeters) and found the robot could swim at nearly 1 meter per second and fly at roughly 6 meters per second, using a similar flapping frequency of about 5 hertz for both.
The surprising part is the transition. Most diving birds paddle at the surface to build up speed before they can take off. The MIT-led team found their robot didn’t need to: by flapping through a steep, roughly 70-degree pitch angle, the FAAV could leap straight out of the water into stable flight. “If you look at birds, most birds need to paddle at the surface to take off,” lead author Raphael Zufferey, an assistant professor of mechanical engineering at MIT and head of the AURA Lab, said in MIT’s announcement of the work. “And the question was, do we need the same for robots? And it turns out we don’t.”
Getting there required solving a design contradiction: the wings had to be flexible enough to minimize resistance while flapping underwater, but stiff enough to keep the robot aloft once airborne — a balance the team tuned through the hydrophobic coating and wing geometry.
A genuinely multi-institutional author list
For a research-administration audience, the mechanics are only half the story. The Science paper’s co-authors are drawn from MIT’s AURA Lab, EPFL (École Polytechnique Fédérale de Lausanne) in Switzerland, and Northwest Indian College, a tribal college based in Bellingham, Washington. That is a genuinely unusual authorship footprint: a US tribal college co-authoring in one of the world’s most selective general-science journals is a rare and concrete example of authorship credit reaching beyond the small cluster of R1 universities that normally dominate Science and Nature bylines. For institutions and funders working on research-equity and broadening-participation goals, this is the kind of outcome — named co-authorship on a flagship paper, not just a diversity statement — that’s worth citing as a model.
The EPFL collaboration also matters in its own right: it reflects the kind of sustained, cross-border lab-to-lab partnership that research offices spend years trying to broker through MOUs and joint funding calls, here showing up as a shared robotics platform and a joint field campaign.
Field trials at Lake Geneva
The team’s field trials for the aerial-aquatic transitions were carried out at Lake Geneva, Switzerland — a logistical detail that underscores the collaboration was not just an author-list formality. Testing a robot that has to survive repeated water impacts, launch reliably, and then fly required physically moving hardware and people to a shared test site, coordinated across the MIT and EPFL teams.
What was funded
MIT’s announcement states the work was supported, in part, by a Marie Skłodowska-Curie Actions fellowship grant — the EU’s flagship early-career researcher mobility funding scheme. MIT’s write-up does not enumerate a full list of funders or grant numbers beyond that fellowship; readers who need the complete funding and acknowledgments record should consult the acknowledgments section of the Science paper directly.
Why it matters for research administrators
Strip away the flapping wings and this is a case study research offices can point to: a project that produced both a genuine engineering advance and a visible, citable instance of cross-border and cross-institution-type collaboration — including a tribal college earning first-tier authorship credit alongside MIT and EPFL. As funders and universities push harder on international partnership and broadening-participation metrics, papers like this one are useful evidence that the goal is achievable in practice, not just in policy language.
Read more: “Leaping out of the water: Aerial-aquatic locomotion with flapping wings,” Science (2026) and MIT’s announcement, “New flapping robot swims and flies like a diving bird” (MIT News, July 9, 2026).
