Flocks of birds sweeping across the sky, schools of fish twisting in the ocean, and swarms of bees buzzing together may seem like simple wonders of nature.
But behind these movements is a hidden intelligence: groups acting in perfect harmony without a leader.
This phenomenon, known as “swarm intelligence,” has long fascinated scientists and inspired engineers who want to recreate it in machines.
If robots or drones could work together in swarms like birds or fish, they might help us solve huge problems—from searching for missing people in remote areas to monitoring the spread of wildfires.
Yet, creating artificial swarms that can coordinate naturally has been one of the toughest challenges in robotics and artificial intelligence.
A new study, published in the journal Proceedings of the National Academy of Sciences, may have found an elegant solution.
An international team of researchers from Radboud University in the Netherlands and New York University has developed a framework that shows how swarms can be controlled using geometry and physics, rather than complicated programming.
“One of the great challenges of designing robotic swarms is finding a decentralized control mechanism,” said Matan Yah Ben Zion, an assistant professor at Radboud University and one of the study’s authors.
“Fish, bees, and birds do this very well—they move together without a leader. Synthetic swarms, by contrast, are clumsy and not yet capable of such large-scale coordination.”
The researchers approached this problem by studying how self-propelled particles behave when influenced by external forces.
They discovered a property they call “curvity”—a natural tendency of particles to bend or curve as they move. Much like electric charges, curvity can be either positive or negative, and it determines whether two particles attract or repel each other.
By applying this idea to robots, the scientists showed that curvity could be encoded directly into a machine’s structure. Robots with certain curvity values would naturally group together, while others would scatter or form flowing patterns. In experiments, pairs of robots with opposite curvity behaved predictably, and the rules scaled up smoothly to thousands of robots.
“This charge-like quantity, which we call curvity, directly controls how robots interact with each other,” explained Ben Zion. “It can cause them to cluster or deflect, just like charges in physics.”
What makes this breakthrough exciting is its simplicity.
The rules are based on basic mechanics, meaning they can be applied to a wide range of technologies, from industrial delivery robots to microscopic medical robots. NYU’s Mathias Casiulis emphasized that this work transforms the problem of controlling swarms into an exercise in materials science, making it more straightforward to design real-world applications.
In the future, swarms of curvity-controlled robots could help deliver medicines within the body, build structures, or respond to natural disasters.
By borrowing nature’s rules and giving them a mathematical twist, scientists are bringing us closer to a world where robots move as gracefully and efficiently as a flock of birds.
Source: New York University
