Engineers have created an incredibly tiny robot that can barely be seen, yet it can “sense, think, and act” on its own.
According to the team, this is the world’s smallest programmable robot capable of moving autonomously through fluid, shrinking previous designs by about 10,000 times.
Researchers at the University of Pennsylvania and the University of Michigan say that until now, no one had managed to put a real computer with a processor, memory, sensors, and a propulsion system into a platform this small.
The robot is smaller than a freckle, no bigger than a grain of salt, and can even balance on the ridge of a fingerprint. It measures just 200 by 300 micrometers wide and 50 micrometers thick. Placed on a penny, it is smaller than the coin’s stamped date, making it almost invisible to the naked eye.
Despite its tiny size, the robot has enormous potential. Its creators explain that the fully programmable device, which works only when submerged in fluid, can move, sense, act, and compute using solar cells that generate about 100 nanowatts of power. It can even measure the temperature of the fluid and communicate that information by performing a small “dance,” similar to how honeybees communicate.
Marc Miskin, a nanorobotics engineer at the University of Pennsylvania, says this is just the beginning. He explains that they have proven it is possible to put a brain, sensor, and motor into something nearly invisible, and have it survive and function for months. This foundation allows the potential to add more intelligence and capabilities, opening the door for a new era of microscale robotics.
Before this, the smallest autonomous programmable robots were over a millimeter in size, a milestone first achieved more than twenty years ago. Shrinking robots further faced challenges because at such tiny scales, the physics changes—forces like drag and viscosity dominate over gravity and inertia. Miskin explains that at this size, moving through water feels like pushing through tar.
The breakthrough came from combining two innovations: a microscopic computer developed by researchers at the University of Michigan and a specially designed propulsion system from the University of Pennsylvania. The robot has no moving parts or limbs, which are difficult to make at this scale. Instead, it moves by generating an electrical field that creates a flow of molecules around its body, making it feel like the robot is in a moving river that it also helps drive.
Fitting a computer onto such a tiny platform required rethinking computer programming and semiconductor circuits, says David Blaauw, a computer scientist at the University of Michigan. The result is a microrobot, five years in development, that can synchronize with others, forming moving groups similar to schools of fish. These groups could theoretically operate autonomously for months if charged by LED light on their solar panels.
The researchers are hopeful that in the future, they can expand the onboard memory to allow more complex programming and more advanced autonomous behaviors. One day, these microscopic robots could even help protect the health of cells in the human body. From such small robots, huge possibilities are emerging. The study was published in Science Robotics.
