Evanston, USA – Engineers at Northwestern University have unveiled the first modular robots with athletic intelligence, capable of running, recovering from damage, and transforming into new shapes in real-world environments.

Called “legged metamachines,” the robots are built from autonomous Lego-like modules. Each module contains its own motor, battery, and computer, allowing it to move independently by rolling, turning, or jumping. When combined, the modules form complex configurations that undulate like seals, bound like lizards, or spring like kangaroos.

“These are the first robots to set foot outdoors after evolving inside of a computer,” said Sam Kriegman, who led the study. “They are rapidly assembled and then quite literally hit the ground running. They can move freely in the wild and easily recover from major injuries that would be fatal to every other wild robot.”

Using AI-driven evolutionary algorithms, the team generated new body configurations not conceived by human designers. The algorithm selected and iteratively refined the best-performing designs, simulating survival of the fittest in a virtual environment. Modular legs could become legs, spines, or tails depending on the configuration, giving the robots extreme versatility.

In outdoor tests, the metamachines traversed gravel, grass, tree roots, sand, and uneven terrain. They could flip upright when overturned, hop over obstacles, spin in the air, and continue functioning even when damaged. Broken modules remain active and can rejoin the robot team.

Kriegman explained, “Metamachines can be rapidly assembled, repaired, redesigned, and recombined. Once assembled, they immediately move themselves across a wide array of unstructured environments.”

The study, published March 6 in the Proceedings of the National Academy of Sciences, marks a major step toward robots that survive and adapt in unpredictable real-world conditions, pointing to a future where machines behave more like resilient, evolving lifeforms than fragile tools.

Podcast: Play in new window | Download