scorecard
  1. Home
  2. tech
  3. Meet The Former NASA Engineer Who Wants To Build A Giant Robot That Can Juggle Cars

Meet The Former NASA Engineer Who Wants To Build A Giant Robot That Can Juggle Cars

Dylan Love   

Meet The Former NASA Engineer Who Wants To Build A Giant Robot That Can Juggle Cars
Tech3 min read

Jul 01, 2014 09:53

Screenshot

If all goes according to plan, former NASA Jet Propulsion Laboratory (JPL) engineer Dan Granett will soon build a 70-foot-tall diesel-powered robot that can use its robo-hands to juggle cars as if they were tennis balls.

A computer-generated demo of what it would look like is above, but just like any good Mythbusters experiment, this needs to first be tested on a smaller scale.

dan ganett

Dan Ganett

Dan Granett

Granett's Kickstarter project seeks to raise $50,000 to build a human-scale prototype of a juggling robot arm. It's a one-armed proof-of-concept design that will be able to handle a 250-pound load, receiving input from a human wearing a remote control "sleeve."

The sleeve detects the movements of a human arm and feeds positioning data to the prototype, which swings into action to perpetually clone the position of the sleeve. The sleeve will also provide "haptic feedback," meaning that while the robot is actually doing the work of juggling an object, the sleeve will provide the feedback of the object's weight as if you were actually holding it.

If the Kickstarter campaign should net $75,000, Granett will build the prototype so that it has two arms, setting it up for some real juggling action. If you want to throw in for his cause, there are BugJuggler stickers, t-shirts, and models to be had, as well as a visit to the BugJuggler machine shop.

bugjuggler

Kickstarter

The haptic feedback sleeve

BUSINESS INSIDER: Why a car-juggling robot?

DAN GRANETT: It all started as a way to bring attention to my nonprofit, StreetPhysics, where we our developing a low-maintenance biowaste-powered turbine for pumping clean water out of the ground in communities too poor to buy fuel.

The underlying theme of the site is to show kids of all ages (including me) how easy it is to build a useful mechanical device once you know how to use a few tools.

Then the juggler took on a life of its own. I had been playing with hydraulics and had done some haptic feedback bench-top experiments where a joystick pushes back on your fingers when the graphic object that you are moving around on the screen bumps into something. Then the phrase "buggy juggler" popped into my head and the stew started to simmer.

BI: Do any NASA-specific skills that carry over into this new undertaking?

DG: I fabricated parts of Shuttle experiments that had to be very weight limited. You learn the limits of strength of materials and where things break so you don't need to overbuild more than just enough to prevent failure. Since childhood, I have been breaking, bouncing, twisting things and have a good "street physics" sense of how materials behave. That gives a headstart on the design after which engineering fine tunes it.

BI: Are you a juggler yourself?

DG: As luck would have it, right next door to our shop is the clown troupe Clowns Not Bombs. [A clown named] Slim Chance there is teaching me to juggle. He says the best way to start is with one hand and two objects. I like bean bags or hacky sacks. They are soft and don't bounce out if you slightly miss. An amusement ride's designer is expected to take the first ride in case of flaws in the design, so I feel that responsibility and should respect the tradition.

BI: What do people need to know about the Kickstarter prototype?

DG: For a small fraction of the cost of the full size machine ($50,000 versus $2.5 million) we can prove the speed and agility of a high-power, high-speed human-machine interface.

Beyond the entertainment aspects, the system will be adaptable in many industrial applications, assistance to the disabled, and emergency rescue operations. The full-size machine will use an off-the-shelf machine vision interface to help position the hand when reaching for the falling car. Our stereo vision system ("eyes" ~2.5 inches apart) has been trained to catch things two feet away so the brain can position the arm based on the angle the eyeballs make when pointed at the object.

BI: Assuming a successful Kickstarter campaign, how long will it take to build a finished prototype?

DG: Our prototype shop in Berkeley has most of the steel and hardware. The campaign will fund the motor, the pricey servo valves, and other electronic and hydraulic components for the haptic control arm. So construction and testing should be accomplished in under 120 days.

Check out the pitch video for the BugJuggler prototype below. If you like what you see, you can give them your money here.

READ MORE ARTICLES ON


Advertisement

Advertisement