Nature has long inspired some of the world’s most advanced robots. From robotic dogs that navigate rough terrain to programmable drones modeled after insects, engineers often look to animals for ideas that solve complex engineering problems.
Now, researchers at MIT and EPFL have taken that concept even further.
They have developed a lightweight robot that can swim underwater, launch itself out of the water, and continue flying through the air, all using flapping wings inspired by diving birds such as loons, puffins, and petrels.
The breakthrough highlights how robotics is increasingly combining biology, mechanics, and artificial intelligence to create machines capable of operating in environments that were once impossible to reach.
For students interested in engineering, robotics, or environmental science, this project offers a fascinating glimpse into the future of intelligent autonomous systems.
Meet the Flapping-Wing Aerial-Aquatic Vehicle (FAAV)
The new robot, called the Flapping-Wing Aerial-Aquatic Vehicle (FAAV), weighs less than 300 grams, roughly half a pound.
Unlike traditional drones or underwater robots, FAAV is designed to move seamlessly between two completely different environments:
- Fly through the air
- Dive into water
- Swim underwater
- Launch back into flight
That transition is far more difficult than it sounds.
Water is approximately 1,000 times denser than air, meaning a robot must deal with completely different forces depending on where it is operating. Most robots are built specifically for one environment or the other.
MIT’s design successfully bridges both.
Learning From Diving Birds
Rather than inventing an entirely new movement system, researchers turned to nature.
Birds such as puffins and loons already solve this challenge every day. They flap efficiently while flying, dive underwater to hunt, then return to the air moments later.
The research team studied data from nearly 100 species of diving birds to understand:
- Wing size
- Flight speed
- Flapping frequency
- Swimming mechanics
- Takeoff angles
They discovered that smaller birds flap their wings much faster than larger birds and that the same wing motion can work surprisingly well in both air and water with the right adjustments.
Instead of copying birds exactly, engineers adapted these principles into a robotic system capable of similar movement.
This approach, known as bio-inspired robotics, has become one of the fastest-growing areas of robotics research.
Engineering a Robot for Two Worlds
FAAV looks surprisingly bird-like.
Its design includes:
- A waterproof central body
- Two flexible flapping wings
- Lightweight waterproof electronics
- A steerable tail
- Replaceable wing sizes for different performance
One of the biggest innovations is the wing design.
The wings are flexible enough to reduce resistance underwater while remaining stiff enough to generate lift during flight.
They are also coated with hydrophobic nanoparticles that repel water, allowing the robot to transition more efficiently from swimming to flying.
Finding that balance between flexibility and strength was one of the project’s biggest engineering challenges.
The Moment of Takeoff
Perhaps the most impressive part of the project is how the robot leaves the water.
During testing, researchers placed the robot about half a meter underwater.
The robot:
- Swam toward the surface
- Tilted upward at roughly 70 degrees
- Continued flapping its wings
- Broke through the surface
- Transitioned directly into flight
Even more surprising, it accomplished this without paddling feet.
Many diving birds rely on their feet to help push themselves off the water before flying.
The MIT robot proved that a carefully designed wing and body system could achieve takeoff without that additional propulsion.
Sometimes engineering doesn’t need to copy nature perfectly. It only needs to understand the underlying principles.
Why This Matters Beyond Robotics
At first glance, this might seem like an exciting laboratory experiment.
In reality, it could lead to entirely new types of autonomous exploration vehicles.
Researchers envision robots like FAAV helping with:
- Marine biology research
- Ocean monitoring
- Coastal erosion studies
- Environmental sampling
- Iceberg observation
- Disaster assessment
- Wildlife monitoring
- Climate research
Instead of sending expensive research vessels, scientists could deploy small autonomous robots that repeatedly collect information from difficult or hazardous locations.
The robot could fly to a location, dive underwater, gather measurements, return to shore, upload its data, and repeat the mission multiple times each day.
This type of persistent data collection could dramatically improve environmental monitoring while reducing costs.
Robotics Is Becoming More Adaptive
Projects like FAAV illustrate an important shift happening across robotics.
Modern robots are no longer built for one specific task.
Instead, they are becoming:
- More autonomous
- More adaptable
- Better at operating across different environments
- Capable of making intelligent decisions during missions
Future robots may fly, swim, climb, drive, or even transition between all of these depending on the situation.
This requires engineers to combine knowledge from many different disciplines, including:
- Mechanical engineering
- Artificial intelligence
- Computer science
- Sensor technology
- Materials science
- Control systems
It is exactly this interdisciplinary thinking that is driving the next generation of robotics innovation.
What Students Can Learn From This
While few classrooms are building aerial-aquatic robots today, the engineering concepts behind FAAV are highly relevant to students learning robotics.
This project demonstrates how robotics involves much more than coding.
Students also explore:
- Engineering design
- Motion and mechanics
- Data collection
- Sensors
- Autonomous navigation
- Testing and iteration
- Real-world problem solving
Perhaps most importantly, students see that inspiration can come from anywhere, including the natural world.
Many of today’s biggest robotics breakthroughs begin with a simple question:
“How does nature solve this problem?”
That curiosity often leads to entirely new technologies.
The Future of Robotics May Look More Like Nature
Bio-inspired robotics continues to grow rapidly because nature has spent millions of years solving complex engineering challenges through evolution.
Whether researchers are studying birds, fish, insects, or animals that walk across difficult terrain, these natural systems offer valuable lessons for designing more capable robots.
MIT’s flapping aerial-aquatic robot is another reminder that tomorrow’s robots may not resemble traditional machines at all.
Instead, they may move with the agility, efficiency, and adaptability found throughout the natural world.
As researchers continue improving designs like FAAV, we can expect to see robots capable of exploring environments that are currently difficult, dangerous, or even impossible for humans to reach.
Bring Real-World Robotics Into the Classroom with LocoRobo
Breakthroughs like MIT’s flapping-wing robot show students what modern robotics can achieve when engineering, programming, and creative problem-solving come together.
LocoRobo helps schools introduce these same concepts through hands-on robotics programs designed for K-12 classrooms. Students build and program wheeled, tracked, legged, and AI-enabled robots while exploring topics such as autonomous navigation, sensors, computer vision, engineering design, and real-world robotics applications.
LocoRobo’s STEM robotics kits include classroom-ready hardware, standards-aligned robotics curriculum, educator training, and implementation support, making it easy for schools to launch or expand engaging STEM and CTE programs.
Whether students are taking their first steps into robotics or preparing for future engineering careers, LocoRobo provides the tools and learning experiences that connect classroom learning with the technologies shaping tomorrow.

































































































































































































