The Record Breaking Robot that Teaches Humans How to Play Table Tennis

Kicking off today, our new weekly video series Meet the Record Breakers – Japan Tour shines a spotlight on some of the most incredible title holders from the nation.

We begin the series, with a futuristic robot that has the uncanny ability of being able to play table tennis better than most humans.

Developed by Omron Corporation, FORPHEUS (Future Omron Robotics Technology for Exploring Possibility of Harmonized aUtomation with Sinic Theoretics) has officially been given the Guinness World Records title for being the First robot table tennis tutor for its unique technological intelligence and educational capabilities.

According to the project’s lead developer Taku Oya, the goal of FORPHEUS was to harmonise humans and robots, by way of teaching the game of table tennis to human players.

The machine is easily able to act as a coach thanks to cutting edge vision and motion sensors it can use to gage movement during a match.

FORPHEUS also features an array of cameras that are situated above the ping pong table which monitors the position of the ball at an impressive rate of 80 times per second.

This functionality also allows the robot to show its human student to see a projected image as to where the return ball will land so that they may improve their skills.

One of the most difficult aspects of the project was determining the algorithms needed for artificial intelligence that would help FORPHEUS determine how well human participants play the game.

However, once the perfect computation was established, it was only a matter of designing a speed-sensing component, which now calculates the opponents ball at one thousand times per second, before the robot was a viable tutor.

The best part of the amazing FORPHEUS? The device actually encourages players to try their best by showing supportive messages along the LED screen situated on the table’s net.

“At the moment it is a human who teaches a robot how to behave or teach,” says Taku, “but in the next 20 years it may be possible that a robot teaches a robot or a robot develops a robot.”

[Via Guinessworldrecords]