A robot that could one day win Wimbledon or the U.S. Open has been developed.
Named ESTHER (Experimental Sport Tennis Wheelchair Robot), the autonomous robo-player can zip around the court and return human shots.
The sporty bot is the brainchild of Georgia Tech’s Matthew Gombolay, associate professor of robotics in the School of Interactive Computing.
He wanted to develop a better training challenge than a stationary ball feeder, one that would act like an always on-call sparring partner or even team up in doubles matches.
The result is ESTHER, a modified wheelchair tennis chair that can race to balls at 10 meters (32 feet) per second and potentially outplay a human.
Matthew Gombolay said: “The wheelchair has the ability to rapidly move around the court and get into position to hit a ground stroke.
“ESTHER’s high-torgue motors can surpass a tennis player’s average side-to-side acceleration, suggesting the potential to out maneuver human opponents in future design iterations.
“ESTHER can move up to two meters after the ball is launched to hit a successful return, which is comparable to the maximum distance moved by human players between 80% of shots.”
The device was named in homage to renowned wheelchair tennis player Esther Vergeer, who held the world No. 1 ranking in women’s wheelchair tennis from 1999 to when she retired in 2013.
Working with more than twenty students, Gombolay authored a paper on building ESTHER, which was accepted to be published in the Institute of Electrical and Electronics Engineers Robotics & Automation Society’s Automation Letters (IEEE RA-L).
The team arranged a network of high-resolution cameras around a tennis court and used computer vision algorithms to help ESTHER recognize an incoming tennis ball.
Using cameras from different angles, they could triangulate where the ball is in space and feed this information to ESTHER.
The team reached a breakthrough when they successfully and consistently programmed ESTHER to locate the tennis ball coming toward it, and to hit a return.
“It took us about two years to get to that point because nobody’s done this before,” Gombolay said. “We built this up from the ground up. Developing that capability was truly exciting.”
ESTHER will now be developed to further simulate the experience of playing against a highly-skilled opponent.
Gombolay added: “ESTHER opens up numerous exciting research opportunities in imitation learning, reinforcement learning, kinodynamic planning, human-robot collaboration and much more.
“What really excites me is that it could be a partner for me one day. It can also be my opponent. It can help me train. I could have it pretend to be the one guy I always lose to because he can exploit this weakness in my game.
“Training against an opponent is psychologically more stressful. Getting closer to simulating real match conditions can help you improve performance.”
Produced in association with SWNS Talker