On November 18, 2026, researchers in the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology published a study detailing a 1.29-gram aerial robot that uses its own sensors to navigate obstacles and land on a sunflower, as reported in Science Robotics.
For the past decade, aerial microrobots built with bumblebee-like agility have remained tethered to laboratory motion-capture systems because they could not carry necessary navigation electronics, Tech Xplore reported. The new flapping-wing machine changes that paradigm by integrating a sensor suite and a miniature microcontroller directly onto its frame. During a thirty-second test flight outside the lab, a human operator directed the robot with a joystick while the machine avoided obstacles and settled on a plant.
This breakthrough positions tiny flying robots for delicate industrial tasks where larger drones fail. Potential applications include monitoring crop health and pollinating blossoms in regions facing declining insect populations, Tech Xplore reported.
“This is a really important milestone for the research community,” said Kevin Chen, an associate professor in the Department of Electrical Engineering and Computer Science at MIT, who heads the Soft and Micro Robotics Laboratory. “This is the first demonstration of an aerial microrobot that is sensory and compute autonomous.”
To achieve this level of autonomy, the MIT team had to increase the payload capacity of their four-winged insect-scale robot without sacrificing speed or maneuverability. Researchers tuned the radius and length of the soft actuators that power the wings, expanding them by 50 percent while boosting net lift forces by 46 percent, Tech Xplore reported.
That optimization yielded a lift-to-weight ratio of 1.9, which easily supports a 244-milligram electronics package.
The package houses an inertial measurement unit for attitude, an optical flow sensor for lateral speed, and a time-of-flight sensor for height measurements, according to wattba.eu.
“When you are talking about a computer which is that small, everything must be thought through carefully,” said Jonathan P. How, a Ford professor of aeronautics and astronautics at MIT.
“There are so many factors to consider—it must do both perception and control, while fitting onto the robot and running fast enough that the vehicle will remain stable during flight.”

During routine test flights, the microrobot hovered and followed point-to-point trajectories with positional errors limited to just a few centimeters. Most calculations during these trials ran on a computer next to the setup, though the robot successfully executed short, autonomous calculations on its own chip during a 10-second hover and a 7-second flight past fixed points, wattba.eu noted.
| Metric | Value |
|---|---|
| Total Robot Weight | 1.29 grams |
| Electronics Package Weight | 244 milligrams |
| Lift-to-Weight Ratio | 1.9 |
| Sensor and Chip Power Draw | 120 milliwatts |
| Total Hover Power | 2.04 watts |
Powering the artificial muscles requires the bulk of the system’s energy, while the onboard sensors and chip consume roughly 6 percent of the total power, wattba.eu reported.
Researchers are currently pursuing two main objectives to prepare the microrobot for field deployment. The team aims to integrate an onboard camera for autonomous object recognition and obstacle avoidance, while also developing power electronics and miniature batteries to enable completely untethered outdoor flight, Tech Xplore reported.
“Small-scale robots are uniquely positioned for these types of tasks, and if this thread of technology becomes successful in the future, it could be very promising,” Chen told Tech Xplore.
Worth a look