Scientists in South Korea have created swarms of tiny, magnetically controlled robots that cooperate similarly to ant colonies. Acting together, these microrobot groups can travel across challenging terrain, move objects much larger than themselves, and even guide organisms.

The research, published December 18 in Device, demonstrates that these microrobot swarms can accomplish tasks too difficult for individual robots. Using a rotating magnetic field for guidance, the robots are able to self-assemble into various shapes. According to lead author Jeong Jae Wie of Hanyang University’s Department of Organic and Nano Engineering in Seoul, South Korea, the robots showed a high degree of adaptability and required minimal direct control when acting together.

In tests, the researchers examined different swarm configurations. One type of swarm climbed over obstacles five times taller than a single robot. Another configuration formed a dense cluster of 1,000 microrobots that floated on water and encircled a pill weighing 2,000 times more than each individual robot, enabling the swarm to move it through a liquid. On dry surfaces, a swarm transported cargo 350 times heavier than a single robot. Other experiments showed that the swarms could clear blockages within tubing, suggesting possible medical applications, and even gently guide the movements of small organisms.

Unlike previous swarm robotics studies that often used spherical robots, this team developed cube-shaped microrobots. Their cube design, each about 600 micrometers tall, increases magnetic attraction through larger contact areas. The robots contain ferromagnetic particles that respond to external magnetic fields. By adjusting how the robots are magnetized, the researchers could control their arrangement and shape.

While these results are encouraging, the system still relies on external magnetic fields and lacks full autonomy, limiting real-world usefulness. Future steps include improving real-time feedback and navigation so that microrobot swarms can operate more independently, potentially enabling applications in environments like arteries or other constrained spaces.