The July 2022 issue of IEEE Spectrum is here!

Close bar

Roboticists around the world are in the process of reverse engineering the anatomical construction of cheetahs and other cats in an attempt to develop faster and more agile legged robots. The latest project, dubbed the Pneupard, hails from Osaka University. Although still early in development, the new biomimetic platform stands out from some of the others through its use of pneumatic artificial muscles as its primary means of locomotion.

The Pneupard is a creation of researchers Andre Rosendo and Shogo Nakatsu, under the supervision of Kenichi Narioka and Professor Koh Hosoda at Osaka University's Graduate School of Information Science and Technology. The team has worked extensively with pneumatic artificial muscles in the past. We've previously covered their biped Athlete, and shared a video of their adorable infant robot Pneuborn.

Pneumatic artificial muscles may be made from a rubber tube sheathed in nylon, but they contract much like the real thing when filled with air. They can pack a lot of power in short bursts and are also highly flexible and impact-resistant, giving them a lifelike quality that is often missing in robots powered by electric motors. The muscle contractions, when combined with limbs that accurately replicate the length, forces, and range of motion of a real cat, naturally generate a lifelike movement.

osaka university pneupard

Rosendo, the project leader, explains that, in the current prototype, each hind limb has eight active muscles, and the fore limbs, which are still under construction, will have six muscles. "The spine will also have muscles, performing flexion and extension," he says. "The main idea of this project is to create a biomimetic platform, where we can replicate feline structure to better understand how they can excel in so many different areas in locomotion. After grasping their secret, it would be possible to apply that knowledge to future robots."

You can see the Pneupard's hind legs walking on a treadmill in the video below. Rosendo notes that the researchers haven't pre-programmed the muscles by hand; the muscles actually activate based on EMG signals recorded from a cat walking on a treadmill. The only sensors used by the robot are force sensors in each foot. "The walking is robust enough to withstand random disturbances, including asymmetry between the right and left caused by the artificial muscle construction method," Rosendo says. "We are compelled to believe that the musculoskeletal structure found in animals is 'intelligent' enough to compensate for disturbances without requiring a sophisticated control method."

For the time being the robot walks with a stilted nervousness, but in the future the team will attach the spine and attempt other gait patterns. They're optimistic, but they're not trying to compete with Boston Dynamics' Cheetah for the world's fastest legged robot—at least, not yet—but their work could pave the way to lighter, less costly robots in the coming years.

More images:

pneupard cheetah robot

pneupard cheetah robot

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman

“I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

This article is part of our special report on AI, “The Great AI Reckoning.”

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓Show less