Roboticists have spent decades taking design cues from nature — building machines that walk like dogs, move like humans, and crawl like insects. A new robot from Duke University just made a case that they’ve been looking in the wrong place.

The robot is called Argus, named after the all-seeing monster of Greek mythology, and the reference is apt. It has 20 legs, no front or back, and looks more like a mechanical sea urchin than anything else. Each telescoping leg radiates outward from a central body with a depth camera at the tip, giving it a nearly 360-degree field of view. Duke researchers published their findings on May 27 in the journal Science Robotics.

The team ran more than 1,500 simulations before settling on Argus’ design. The question they were working from wasn’t which animal to copy, but how symmetrical a machine could be in every direction—a concept called dynamic isotropy. It’s scored on a scale of 0 to 1, measuring how evenly a robot can move its center of mass regardless of which way it’s facing. Most advanced robots today score below 0.6. Argus scored a 0.91.

The key was building the robot’s body around a regular dodecahedron, a three-dimensional shape with 12 pentagonal faces. That geometry gives Argus an almost uniform field of view and lets it move in any direction without stopping to reorient itself first.

“When a robot can accelerate equally well in every direction, it stops needing to face the world in any particular way,” said Boyuan Chen, director of Duke’s General Robotics Lab and co-author of the study. “Forward and backward become the same. Left and right become the same. The whole problem of robot control changes character.”

In footage from the Duke campus, Argus crossed concrete, grass, sand, wet surfaces, and bark. It cleared obstacles up to 5 inches tall, climbed walls, moved through trees, and kept going after three of its legs were broken. For good measure, it also pushed a 3-foot cube across a surface while rolling.

“The first time we saw it navigate among trees and rough terrain, even under heavy collisions, we knew this was something different,” said Jiaxun Liu, a doctoral student and co-author on the paper.

The researchers are clear that Argus is a proof of concept. The actual contribution is the methodology: a mathematical approach to evaluating robot bodies that could change how engineers design new ones from the ground up.