Built to Take a Hit: Tiny Flying Robot is a Step Closer to Flight Beyond the Lab

Insect-sized robots that can navigate natural environments could someday assess disaster areas, aid in search and rescue and pollinate plants in coordinated swarms. However, recent efforts have barely left the comfort of their research homes due to their fragility.  

In a new paper in Science Robotics, Wenzhong Yan, an assistant professor of mechanical and aerospace engineering at the University of California, Davis, and his collaborators from UCLA and MIT report a lightweight, soft flying robot that can generate enough lift to take off independently. And the robot can do those things while surviving impacts and compression.  

“These types of robots have only lived in laboratories because they are so fragile,” he said. “Now, with this kind of very resilient capability, they actually can survive in the environment.”  

Flexing a New Material 

Many insect-scale micro-aerial vehicles, or MAVs, rely on rigid wings, transmissions and actuators. During a collision, these components are fragile and vulnerable to damage because they cannot bend to absorb and distribute impact. An insect’s body, on the other hand, contains flexible structures that can.   

Yan and his collaborators used a polymer film to design an MAV weighing 185 milligrams — about as heavy as four drops of water. Its cone-shaped wings are designed to generate lift and are powered by an electrostrictive actuator.  

“I said, ‘This is a really cool and powerful material. We shouldn’t waste its potential,’” Yan said. “‘It’s two-dimensional and compliant, which makes it very difficult to break under compression.’” 

The actuator comprises layers of polymer that bend when a voltage causes one layer to expand. Rapidly switching the voltage on and off creates the flapping motion needed for takeoff.  

The soft wings partially fold during one half of each wingbeat and reopen during the other, creating the difference in air resistance needed to produce upward lift. Their compliant structure also allows them to deform under pressure without breaking. 

Testing Resilience 

To assess the prototype’s resilience and its behavior in harsh surroundings, Yan and his collaborators subjected it to rigorous testing. It was directed to collide with an obstacle, hit by a flyswatter and flattened by aluminum blocks weighing 11,000 times its own weight.  

The prototype recovered in every instance.  

“The novel material and structure allow the robot to survive a sudden impact or high compressive load and also generate sufficient aerodynamic lift, which is unlike many other micro robots,” Yan said. “This design embodies what I call ‘mechanical intelligence.’” 

Building the Next Iteration 

Now that he has demonstrated a flexible, resilient MAV capable of takeoff while connected to an external power supply, Yan will work on controlling flight maneuvers, including turning and landing, and increasing the MAV’s payload capacity.  

The current prototype receives power via external wires. To make the design applicable in the real world, future versions would need to include power and control systems, as well as sensors or cameras. 

Over the long term, Yan sees considerable potential for this type of technology to be sent into the wild.   

“These robots could be used for environmental monitoring, agricultural pollinating and search and rescue missions in disaster areas. Because they can be resilient, withstand collisions and be compressed, they can survive all these kinds of mechanical interactions with the environment.” 

Yan worked with several members of UCLA’s Soft Materials Research Laboratory, including Yuan Zhu, Hanxiang Wu, Dawei Sun, William Budiman and Kede Liu. The lab is led by Qibing Pei, a professor of materials science and engineering and mechanical and aerospace engineering at UCLA, who also collaborated on the project. Yufeng “Kevin” Chen, an associate professor of electrical engineering at MIT and an expert on micro robots, contributed to the research. 

Support for this research came from the Office of Naval Research, the National Science Foundation, the UCLA Academic Senate Research Allowance Program and the UC Davis College of Engineering.   

Read the paper in Science Robotics