CNET también está disponible en español.

Ir a español

Don't show this again

T-Mobile CEO to step down 2021 Ford Mustang Mach-E electric SUV The Mandalorian Walmart Black Friday 2019 Google Stadia Early Black Friday Deals

Parkour robot takes on uneven terrain

RHex is a six-legged all-terrain robot that can run, leap, backflip and climb.

An older version of RHex.
(Credit: DARPA)

RHex is a six-legged all-terrain robot that can run, leap, backflip and climb.

One of the problems with robots is that automated movement can cause hiccups when it reaches a change in terrain. It's something that US military research agency DARPA has been looking at trying to solve for a while with a variety of robots, usually somewhere on the spectrum of "terrifying".

RHex (pronounced ("Rex") is a little different. The small robot is described as an "over-excited puppy" and, insofar as a robot can be, is of the rather cuter variety.

Thanks to its six springy legs, RHex is capable of a very nifty trick: navigating very uneven terrain. Inspired by parkour, the robot is being programmed by graduate student Aaron Johnson and professor Daniel Koditschek, both of the Department of Electrical and Systems Engineering in the School of Engineering and Applied Science of Koditschek's Kod*Lab in the University of Pennsylvania's General Robotics, Automation, Sensing and Perception (GRASP) Lab, to use its single-jointed limbs to go almost anywhere.

This isn't the first iteration of RHex. Its development, funded by DARPA, began in 1998, and the robot has been a collaboration between several US universities. This version, known as XRL, or X-RHex Lite, is both lighter and more agile than previous versions.

With its curved legs, RHex can activate different sequences to perform double-jumps, flips and even pull-ups: the robot can launch itself into a vertical position, hook its legs over the surface of the object it is trying to climb, and pull its body up. The aim, of course, is to create a robot that can provide services in rough areas: environmental monitoring, for example, or search and rescue.

The project is still ongoing, but Johnson and Kodischek recently presented the paper of their research so far, "Toward a Vocabulary of Legged Leaping",at the IEEE International Conference in May, where it was selected as a finalist for the best student paper.