"Intrinsic Dynamics" System Delivers "Exceptionally Effortless" Robot Locomotion
Researchers propose minimal-energy movement approach inspired by bouncing balls, swinging children, and horses trotting.
Researchers from the Technical University of Munich (TUM) and the German Aerospace Center (DLR) have come up with a way to make robots' movements near-effortless — by putting a little natural shimmy in their sashay.
"Activities like ball bouncing and trampoline jumping showcase the human ability to intuitively tune to system dynamics and excite motions that the system prefers intrinsically," the researchers explain by way of background to their work. "This human sensitivity to resonance has been experimentally supported for interactions with simple linear systems but remains a challenge to validate in more complex scenarios where non-linear dynamics cannot be predicted analytically."
"For the first time, we have succeeded in making these intrinsic, highly efficient movements calculable," says project lead Alin Albu-Schäffer, TUM professor, of the team's breakthrough. "The tool makes it possible to find out which movements of a system are particularly economical."
"You can compare it to a child sitting on a playground swing and receiving an energy impulse at the highest point from the parent who is pushing," adds first author Annika Schmidt. "[Only] humans don't need a lot of equations in their heads to time their push exactly — they do it intuitively."
Giving that same ability to robots is key to improving their efficiency, the researchers say, in a similar way to how a four-legged animal will naturally fall into a trot as it picks up speed. Known as "intrinsic dynamics," delivering the same tools to roboticists means better efficiency — as proven with BERT, a quadrupedal robot that was given six new "exceptionally effortless" movements identified in the research. These movements, based on natural oscillation, are extremely low energy — to the point, the team claims, that they would require no energy at all in a world without friction.
"We have thus confirmed the hypothesis that efficient gaits can be realized by exploiting natural oscillation patterns," Albu-Schäffer explains. In experiments, the BERT equipped with the ability to exploit intrinsic dynamics moved faster and more dynamically than identical robots programmed with conventional movement patterns — even extending to better jumping and hopping skills.
The team's work has been published under open-access terms in the journal PLOS Computational Biology.
Main article image courtesy of Andreas Heddergott/TUM.