Ball catching robot, 80% accuracy in 5 milliseconds (w/ video)

Apr 29, 2011 by Katie Gatto weblog

(PhysOrg.com) -- DLR, an aerospace agency based in German, has modified its flagship robot, known as the Rollin’ Justin, in order to make it into a lean, mean, catching machine. The Rollin’ Justin's modifications allow the machine to catch balls that are thrown in its direction with an accuracy rate of about 80%. The Rollin’ Justin's robots accuracy rating is better than the average uncoordinated human. It is certainly better than this reporters accuracy rate. It is not as good as say, a Major League Baseball player, but then again most of us cannot say that we are as good as the pros either. Considering they also have near unlimited stamina, at least until the battery runs out, it may be a great partner for the standard game of catch, or for the position behind the batter.

This video is not supported by your browser at this time.

The machine can actually catch up to two balls at once, and while this may not sound like a lot on paper, but just try doing it in the back yard. It is not exactly an easy task. The will, when the balls are released, calculate the flight path of those two balls, and then position its hands into a catching position that is within 2 centimeters of the point where the ball is expected to be, in about just 5 milliseconds. The system does work with the help of external computers. They are needed in order to handle the processing and make the calculations possible in real time.

The Rollin’ Justin weighs about 45kg , and it is equipped with a 3D camera system that allows it to pinpoint the objects distance, which needed in order make the calculation. The robot is built with 43 joints, with 5 in his torso, 7 in each of the arms, and 12 in each hand. It is also equipped with a total of 84 sensors.

Explore further: Students turn $250 wheelchair into geo-positioning robot

More information: elib.dlr.de/67146/
Via IEEE

Related Stories

Invention breathes new life into tennis balls

Jun 14, 2005

The traditional cry of "new balls please" at tennis courts throughout the country could become a thing of the past thanks to a new invention by a University of Bath student. Aimée Cubitt, a final year Mechanical Engineering stu ...

Recommended for you

Students turn $250 wheelchair into geo-positioning robot

6 hours ago

Talk about your Craigslist finds! A team of student employees at The University of Alabama in Huntsville's Systems Management and Production Center (SMAP) combined inspiration with innovation to make a $250 ...

Using robots to study evolution

Apr 14, 2014

A new paper by OIST's Neural Computation Unit has demonstrated the usefulness of robots in studying evolution. Published in PLOS ONE, Stefan Elfwing, a researcher in Professor Kenji Doya's Unit, has succes ...

User comments : 14

Adjust slider to filter visible comments by rank

Display comments: newest first

Isaacsname
not rated yet Apr 29, 2011
Ninja robot soon to follow...I'm sure of this.
trekgeek1
5 / 5 (3) Apr 29, 2011
Video didn't work for me. Try this link

http://www.youtub...PwP3s7s4
Isaacsname
not rated yet Apr 29, 2011
Video didn't work for me. Try this link

http://www.youtub...PwP3s7s4


They should get together with these researchers

http://www.youtub...jVlaLBmk
antialias
5 / 5 (2) Apr 29, 2011
Next they'll be catching bullets.
SmartK8
not rated yet Apr 29, 2011
antialias: In the WWIII, they will. ;)
Starbound
not rated yet Apr 29, 2011
Great videos, gentlemen! Kind of makes me want to make a robot.
Fig1024
5 / 5 (1) Apr 29, 2011
can an an electric motor joint achieve speeds necessary for bullet catching? even if it was possible to calculate bullet position and trajectory instantly, what kind of motors would handle the arms?
Deadbolt
3 / 5 (4) Apr 29, 2011
Humans average about 215ms for visual reaction, with a limit towards 100. A housefly, because of its tiny optical nerve distance and relatively limited processing, can manage 20 milliseconds before responding to stimuli. At 5ms, This machine thinks 4 times quicker than a fly apparently.

With some modifications, this machine COULD potentially block bullets, even with just human speed. If it only has to move an arm 50cm to block the bullet, and it can spot a 450m/s bullet at 2.25 meters out the barrel, then, at a perfectly achievable arm speed of 20m/s, it could grab the bullet when fired from 13.5 meters (11.25 meters plus reacting/calculating time) away...

Now, why you'd want to "grab a bullet" when it's the equivalent of getting shot in the hand, I don't know. Future death robots will be better off just dodging.

Actually... robot bodyguards, anyone?
VOR
not rated yet Apr 30, 2011
cool stuff. will be even more impressive when it throws them back.
After all, playing catch is so much more fun.
Beard
not rated yet May 01, 2011
The bullet would disintegrate in your grasp even if you caught it. The robots would need hands/shields made of a hardened material immune to bullet impacts. Then they could block bullets.

Strap one of those badboys on your back and laugh as the terrorists try to shoot you.
trekgeek1
not rated yet May 01, 2011
The robot could just tap the bullet to send it off course. This way most of the kinetic energy isn't transferred to the hand.
Decimatus
not rated yet May 01, 2011
More likely the robot will kill/disarm the gunman before he can even tell his little finger to pull the trigger.

I wonder how fast robot legs will be able to get to 60 mph?

"It's ok Chris. I am pretty sure she would have crushed your scrote into a Diamond with her robot strength arm." - Stewie
rwinners
not rated yet May 01, 2011
Cool. Now rich dads can buy a standin.
emsquared
not rated yet May 04, 2011
Now if they can get 'em to throw and swing a bat, we'd have BASE WARS in no time!
(http://en.wikiped...se_Wars)
Rollin' Justin actually kinda looks like the one batting... wierd.

More news stories

Quantenna promises 10-gigabit Wi-Fi by next year

(Phys.org) —Quantenna Communications has announced that it has plans for releasing a chipset that will be capable of delivering 10Gbps WiFi to/from routers, bridges and computers by sometime next year. ...

Unlocking secrets of new solar material

(Phys.org) —A new solar material that has the same crystal structure as a mineral first found in the Ural Mountains in 1839 is shooting up the efficiency charts faster than almost anything researchers have ...

Floating nuclear plants could ride out tsunamis

When an earthquake and tsunami struck the Fukushima Daiichi nuclear plant complex in 2011, neither the quake nor the inundation caused the ensuing contamination. Rather, it was the aftereffects—specifically, ...

Patent talk: Google sharpens contact lens vision

(Phys.org) —A report from Patent Bolt brings us one step closer to what Google may have in mind in developing smart contact lenses. According to the discussion Google is interested in the concept of contact ...

Meteorites yield clues to Martian early atmosphere

(Phys.org) —Geologists who analyzed 40 meteorites that fell to Earth from Mars unlocked secrets of the Martian atmosphere hidden in the chemical signatures of these ancient rocks. Their study, published ...