Robots Learn How to Play Catch With Soulless, Mechanical Precision

Today's Best Tech Deals

Picked by PCWorld's Editors

Top Deals On Great Products

Picked by Techconnect's Editors

Last year, the Institute of Robotics and Mechatronics at the German Aerospace Center created the "Rollin' Justin" robot, a technical marvel that could catch a ball through a mix of precision, user input, and motion sensors. It proved to be a success, but there's only so much research you can do with a single-function robot. Enter "Agile Justin," the counterpart machine that can pitch a ball with a great deal of finesse.

Huffington Post relayed video footage of the two robots in action, as each unit showed off their respective skills in a short game of catch. It's a neat thing to watch in terms of mechanical detail, especially when you consider that each robot has to mimic the hand-to-eye coordination required for a seemingly simple thing. Researcher Berthold Baumi breaks it down in more technical detail in his write-up of the team's "Justin" programs.

Overall, it's pretty impressive to see in motion. Luckily, Justin and Justin aren't equipped with any degree of artificial intelligence, because they could easily be throwing lethal objects through the air at unsuspecting humans. (Hey, it could happen.)

Headmounted HD stereocameras track the thrown balls. The rapid catching motion introduces vibrations to the system, which the control can not cancel completely. Therefore, a supporting inertial measurement unit is used for predicting the ball's trajectory.

During the flight the predictions continuously improve and a realtime path planner decides where, when, and in which configuration to kinematically optimal catch the ball. For this, a non- linear optimization problem (including simple collision avoidance) is repeatedly solved on an external computing cluster (32 cores).

Either way, it seems like the next logical step will be a third "Justin" robot that can pitch and catch a ball in one fluid motion. It's all about baby steps, though. Maybe one day, all the different robotics teams out there will figure out how to make a fully functional multi-use humanoid machine that can play a full game of baseball.

[Huffington Post; Image via Hizook]

McKinley Noble is a former GamePro staff editor, current technology nerd and eternal mixed martial arts enthusiast. He also likes Japanese sports dramas and soap operas. Follow him on Twitter or just Google his name.

Like this? You might also enjoy...

Get more GeekTech: Twitter - Facebook - RSS | Tip us off

Note: When you purchase something after clicking links in our articles, we may earn a small commission. Read our affiliate link policy for more details.
Shop Tech Products at Amazon