Researchers at the University of Pennsylvania and the University of Michigan have achieved a major breakthrough by creating what they describe as the world’s smallest fully programmable, autonomous robots. These microrobots are microscopic swimming machines, each measuring roughly 200 by 300 by 50 micrometers—smaller than a grain of salt. Despite their tiny size, they can sense and respond to their environment, operate continuously for months, and be manufactured for about one cent each. Working at the same scale as many biological microorganisms, they open new possibilities in medicine, manufacturing, and fundamental research.
A parallel advance has also been reported by researchers at the Chinese Academy of Sciences, who demonstrated multi-material microrobots capable of grasping objects as small as individual cells. Using femtosecond laser direct writing, they integrated different materials at the micrometer scale to create hand-shaped devices that can grab, carry, and release microscopic targets. Together, these developments show that microrobots are rapidly evolving from simple experimental systems into versatile, functional machines.
One of the most significant challenges the Penn team faced was overcoming the physics of the microscopic world. At human scales, gravity and inertia dominate motion, but at cellular scales, viscosity and drag take over, making water feel more like thick syrup. Traditional robotic strategies using limbs or joints fail under these conditions. To address this, the researchers designed a novel propulsion system based on electrokinetic effects. Instead of moving parts, the robots generate electric fields that push ions in the surrounding fluid, which in turn move nearby water molecules. This allows the robots to “swim” smoothly, maneuver in complex patterns, and even coordinate their movement in groups, much like schools of fish.
Durability and longevity are key advantages of this approach. With no moving mechanical components, the robots are resistant to damage and can be transferred between samples without breaking. Powered by tiny solar cells and charged by simple LED light, they can operate autonomously for months. This makes microrobots practical not just as laboratory curiosities, but as platforms for long-term experiments and applications.
The University of Michigan team provided the final piece of the puzzle by integrating an onboard computer. Achieving true autonomy required fitting a processor, memory, sensors, and power management onto a chip far smaller than a millimeter, running on just 75 nanowatts of power. To do this, the researchers designed ultra-low-voltage circuits and radically simplified computer instructions so entire programs could fit into extremely limited memory. The result is the first sub-millimeter robot with genuine onboard intelligence.
These robots can sense temperature with high precision, respond to environmental changes, and even communicate their measurements through distinctive movement patterns that researchers decode visually—an approach reminiscent of honeybee communication. Each robot can be individually programmed using pulses of light, allowing large groups to perform coordinated but distinct tasks.
According to the researchers, this achievement is only the beginning. The current system is a general platform that can be expanded with new sensors, faster motion, and more complex behaviors. By proving that microrobots can combine propulsion, sensing, computation, and durability at this scale, the work opens the door to a new era of robotics operating in environments once thought inaccessible.
www.therobotreport.com/researchers-create-programmable-autonomous-robots-smaller-than-a-grain-of-rice/