Student Robert Gregg earns ACC’s Hugo Schuck Award for robotics work

4/29/2009 Kim Gudeman, Coordinated Science Lab

ECE graduate student Robert Gregg has been awarded the 2009 Hugo Schuck Award by the American Control Council (ACC) for his work on control methods for dynamic walking robots. Gregg won the best student paper for the same research at last year’s ACC.

Written by Kim Gudeman, Coordinated Science Lab

ECE grad student Robert Gregg and Adjunct Professor Mark Spong have developed robots that walk more like people by decomposing motion in humans’ sagittal plane.
ECE grad student Robert Gregg and Adjunct Professor Mark Spong have developed robots that walk more like people by decomposing motion in humans’ sagittal plane.

ECE graduate student Robert Gregg has been awarded the 2009 Hugo Schuck Award by the American Control Council (ACC) for his work on control methods for dynamic walking robots. Gregg won the best student paper for the same research at last year’s ACC.

His paper, “Reduction-based Control with Application to Three-Dimensional Bipedal Walking Robots,” unveils a new way of controlling robots to mimic human walking. The paper was co-authored by ECE Adjunct Professor Mark W Spong.

“Humans walk more efficiently than any other land-based species,” says Gregg, who performs his research in the Coordinated Science Lab. “As robots become more advanced, they will have to mimic human walking in order to interact with humans, navigate more versatile environments, and use energy efficiently.”

Humans use passive dynamics to walk. In other words, people allow gravity to perform a lot of the work, which minimizes energy and allows humans to walk farther than four-legged species. In contrast, many contemporary bipedal robots such as the Honda ASIMO are controlled at every point of movement. They can stop mid-gait and not fall over, which requires huge amounts of energy.

Robert Gregg
Robert Gregg

Gregg and Spong have applied the same passive dynamics principles to bipedal walking robots. Of the three planes-of-motion that humans move in, most movement occurs in the sagittal plane, where Gregg and Spong have focused their work. They have decomposed this fundamental motion using geometric reduction-based control theory, allowing them to build passive dynamic walking gaits in three dimensions. The result is a walking robot that can navigate desired paths while “falling” from one step to the next as a human does.

Gregg will receive the honor at the ACC’s conference in St Louis (June 10-12). The council gives a Schuck Award in two categories, application and theory. Gregg’s paper won in the theory category.

“I’m glad that people are starting to pay more attention to humanoid robots and how we move in everyday life,” he said.

For more information about Gregg’s research, please visit http://decision.csl.illinois.edu/~rgregg.


Share this story

This story was published April 29, 2009.