Bee-sized robots can not only fly and hover, but now they have learned to land gracefully. Harvard University’s RoboBee recently upgraded its “landing skills”, inspired by the “big mosquitoes” you may often see in summer! Scientists use the clever design of nature to make this micro robot more stable and less damaged, and it is one step closer to the goal of truly “free flight”. What are the tricks behind this? Let’s take a look at the story of RoboBee.
Harvard University’s RoboBee has long demonstrated its ability to fly, dive and hover like a real insect. But without a safe way to land, these aerial wonders are difficult to play their true value.
As one of the landmark projects of Harvard’s Microrobotics Laboratory, RoboBee has recently been equipped with the most reliable landing gear ever – inspired by the most elegant “landing expert” in nature: the crane mosquito.
The team, led by Robert Wood, a professor at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS), published their findings in the journal Science Robotics. They installed a pair of long jointed legs for RoboBee to help it land smoothly from the air, and upgraded the controller system to make it decelerate and cushion better during landing, achieving a “soft touchdown.”
An important goal of these improvements is to protect the precision piezoelectric actuators that RoboBee relies on – these “muscles” for flight are efficient but extremely fragile and can be easily damaged by hard landings or collisions.
Landing RoboBee has always been a technical challenge, in part because it is extremely light – weighing less than 0.1 grams and with a wingspan of only 3 cm. Early versions often suffered from the “ground effect” when approaching the ground: the airflow generated by flapping wings disturbed its own stability, just like the downwash of helicopters when landing.
“The way we landed it before was to simply cut off the power a little way off the ground and ‘pray’ it landed safely,” recalls Christian Chan, a graduate student who led the mechanical redesign and co-first author of the project.
To solve this problem, the research team upgraded the robot’s “brain” – the controller – to enable it to recognize and adapt to ground effect. The work was led by co-first author Nak-seung Patrick Hyun, a former postdoctoral researcher, who led the robot’s controlled landing tests on leaves and hard ground.
“The key to a successful landing of an aircraft is to slow down as much as possible before contact with the ground and dissipate energy quickly after contact,” said Hyun, now an assistant professor at Purdue University. “Even though RoboBee is small, the ground effect is still significant when flying close to the ground, and the bounce and tumbling after landing will also increase instability.”
To solve this problem, the team looked to nature for “design inspiration.” They found an insect that is often mistaken for a giant mosquito – the crane fly, a large insect known for its slow and stable flight ability.
“The size of our platform is very close to that of a crane fly,” Chan said, “and its long jointed legs seem to be designed to cushion landings.”
The team consulted records of insect specimens in Harvard University’s Museum of Comparative Zoology and designed a variety of robotic leg prototypes based on the crane fly’s leg structure. In the end, they chose a multi-joint design similar to that of the crane fly and used the precision manufacturing process developed by Harvard’s Microrobotics Laboratory to adjust the stiffness and damping of each joint.
During this process, insect movement expert and postdoctoral researcher Alyssa Hernandez also joined the research. She studied for a doctorate in the Department of Organismic and Evolutionary Biology at Harvard University and brought a wealth of knowledge about insect movement to the project. “RoboBee is an ideal platform for exploring the integration of biology and robotics,” she said. “We can draw inspiration from countless insects in nature, and also use robotic platforms to reverse-verify hypotheses about biological movement.”
Currently, RoboBee still needs to be connected to an external control system via a tether. The research team’s long-term goal is to create a fully autonomous micro-flying robot, which requires breaking through the three major technical bottlenecks of sensing, power supply and control at the same time.
“Full autonomy is our holy grail goal,” said Professor Wood, “but before we can achieve it, we still need safety tethers to avoid electrical and mechanical failures. To get rid of these tethers, the first thing to solve is how to land safely.”
Micro robots such as RoboBee are not only amazing in technology, but also have a wide range of applications – from disaster monitoring to environmental detection, and even future agricultural automation. Chan specifically mentioned that one of the applications he is most looking forward to is artificial pollination: in the future, perhaps we can see a group of RoboBees flying around vertical farms, completing pollination work for plants, becoming a true “robot bee”.
Reference: https://www.sciencedaily.com/releases/2025/04/250416151924.htm
