Mobile-Menu

Put into action Four-legged robot learns to walk energy-efficiently

Source: TU Munich | Translated by AI 2 min Reading Time

Related Vendors

TU München
FAULHABER Drive Systems
Taiwan Machine Tool & Accessory Builders' Association (TMBA)
Wuxi InfiMotion Technology Co., Ltd.

When four-legged creatures speed up their pace, they automatically start trotting as it saves energy. A robot from TU Munich can now do this too.

This is the four-legged robotic dog Bert. With him, researchers at TU Munich have now achieved a breakthrough in insights into the most energy-efficient modes of locomotion ...(Image: TU Munich / A. Heddergott)
This is the four-legged robotic dog Bert. With him, researchers at TU Munich have now achieved a breakthrough in insights into the most energy-efficient modes of locomotion ...
(Image: TU Munich / A. Heddergott)

It has been known for over 40 years that quadrupeds transition to energy-saving trot at a certain running speed, say researchers from the Technical University of Munich (TUM). Now, with regard to sensor-based robotic systems and "intelligent" assistance systems, it has been possible to transfer this behavior to the movement of robots. Experts refer to intrinsic dynamics, which humans and animals use for their energy-efficient movements. They adapt the stiffness of their muscles, for example, when running on harder surfaces. These instinctive adjustments happen automatically and are difficult to identify, it is said. This applies both to humans and to complex robotic systems.

On the track of the most economical movement with Bert

However, with a new tool from TUM, the knowledge gap has now been closed. This makes it possible to determine which movements of a system are particularly economical. An important test object is Bert, a four-legged robot that looks like a small dog. Bert was designed at the German Aerospace Center (DLR). This research focused on "efficient and versatile bipedal locomotion" and is funded by an ERC Advanced Grant from the EU. The involved experts identified six movement patterns for the mechanical quadruped that require no energy in a frictionless world, as emphasized. They partially correspond to known gaits of quadrupeds (walking, trotting, or hopping). This has confirmed the hypothesis that efficient gaits could be achieved by utilizing natural oscillation patterns.

Robot race confirms the hypothesis in practice

To implement these movements in a normal system (i.e., with friction), a computer-controlled controller has now been implemented to give an impulse at the right moment, it says. You can imagine it similarly to a child sitting on a swing and receiving an energy impulse from their parents at the highest point to push them. But there is a difference! A human does not need a lot of equations in their head for exact timing—they do it intuitively. However, the robot must be taught the right rhythm. A race between three Bert models revealed the correctness of the hypothesis. The robot dog that mastered the method of intrinsic motion moved forward more dynamically and quickly, hopping rather than in the conventional programmed movement patterns of its siblings.

Subscribe to the newsletter now

Don't Miss out on Our Best Content

By clicking on „Subscribe to Newsletter“ I agree to the processing and use of my data according to the consent form (please expand for details) and accept the Terms of Use. For more information, please see our Privacy Policy. The consent declaration relates, among other things, to the sending of editorial newsletters by email and to data matching for marketing purposes with selected advertising partners (e.g., LinkedIn, Google, Meta)

Unfold for details of your consent

Related content