Sign in with
Sign up | Sign in

Nvidia's Tegra Platform May Be Heading to the Moon

By - Source: Softpedia | B 24 comments

Nvidia's Tegra platform is being used to design two rovers that will travel across the moon next year.

Nvidia's Tegra platform may be heading to the moon next year as part of Google's Lunar X-PRIZE competition. One of the competitors, a privateer team named Team Synergy Moon, plans to use several Tegra systems to create a couple of rovers capable of crawling across the lunar landscape.

So what is Google's Lunar X PRIZE? It's a $30 million competition for the first privately funded team to send a robot to the moon.

"A total of $30 million in prizes are available to the first privately funded teams to safely land a robot on the surface of the Moon, have that robot travel 500 meters over the lunar surface, and send video, images and data back to the Earth," the Lunar X PRIZE site explains. "Teams must be at least 90-percent privately funded, though commercially reasonable sales to government customers are allowed without limit."

To develop their robots, Team Synergy Moon collaborated with Martin Peniak, a young scientist known for his work on the Mars Rover Simulator developed in conjunction with the European Space Agency (ESA). Peniak reportedly chose to take the Tegra route after reading an article about a neural network using Nvidia GPU acceleration. After that, Peniak chose CUDA as a basis for the rovers' development and underlying AI.

As seen on Team Synergy Moon's website, the Tesla Prospector will be the one that fulfills the competition's 500 meter requirement while taking samples and sending images and data back to earth. The team's other robot, the Tesla Surveyor, will be a bit more observant, packing twin HD cameras while rolling across the lunar surface like a silver marble -- a lunar tour guide of sorts.

"Both rovers feature an internet based control system that will allow tourists, scientist and prospectors to take control of the rovers for a virtual excursion on the moon," the team reports. "Both the Telsa Prospector and Surveyor will be capable of autonomous operations, able to take care of themselves and carry out exploration missions on their own, saving their observations for later transmission to earth or streaming near real time video of their adventures. The Tesla Prospector will also carry a set of microbots, that will carry out specialized scientific and artistic projects."

If Team Synergy Moon wins the competition, the rocket sending a manned orbiter -- along with a moon lander containing the two rovers -- will launch in December of 2012.

Google Lunar X Prize - TEAM SYNERGY MOON - Tesla Surveyor Rover

Display 24 Comments.
This thread is closed for comments
  • 9 Hide
    sunflier , December 7, 2011 11:40 AM
    Quote:
    That's no moon. It's a space station.
  • -5 Hide
    BruceOTB , December 7, 2011 11:46 AM
    What if we just quarry the moon?
  • 4 Hide
    jamie_1318 , December 7, 2011 11:51 AM
    I don't think you can use Tesla in space. That CPU is way too sensitive to cosmic rays, like any other CPU designed to work in space. They use custom designed CPUs in anything that goes to space in order to avoid being totally fried by space-radiation. If anyone has any more information on the topic I'd be interested to hear.
  • 4 Hide
    songemu , December 7, 2011 12:09 PM
    I have a sneaking suspicion that Tegra hardware is not radiation-hardened... unless this rover is going to be made of lead and clay, he's in for a world of hurt. Processors in satellite applications operate at a couple hundred megahertz at best because of radiation.
  • 1 Hide
    jabliese , December 7, 2011 1:13 PM
    The usual cheap solution is to have 3 or more CPU's working on the same problem. If a majority come up with the same answer, go with it. If you want to be a real stickler, make sure all CPU's agree before using the answer. In case of disagreement, recalculate. And, yea, it will probably be underclocked.
  • 4 Hide
    gmarsack , December 7, 2011 1:52 PM
    My CPU is a neural net processor; a learning computer.
  • 2 Hide
    cookoy , December 7, 2011 2:17 PM
    the rovers will go to the moon; the tegra platform used to design the rovers will stay on earth. the rovers will have internet based control system. is the internet available now on the moon?
  • -4 Hide
    iceman1992 , December 7, 2011 2:18 PM
    jamie_1318I don't think you can use Tesla in space. That CPU is way too sensitive to cosmic rays, like any other CPU designed to work in space. They use custom designed CPUs in anything that goes to space in order to avoid being totally fried by space-radiation. If anyone has any more information on the topic I'd be interested to hear.

    I believe they're using Tegra :) 
  • 0 Hide
    dark_knight33 , December 7, 2011 3:07 PM
    BruceOTBWhat if we just quarry the moon?



    Ever seen "The Time Machine" (2002)? Not to mention the moon is mostly iron, so far as we know, and there is no shortage of iron on Earth, nor is it economically feasible to transport any material from the moon for commercial sale.
  • 0 Hide
    LuckyDucky7 , December 7, 2011 5:08 PM
    Quote:
    Processors in satellite applications operate at a couple hundred megahertz at best because of radiation.


    They also use really big (1000 nm) transistors.

    You see... you don't actually need anything more powerful than a 386 for space computing (not only that but it's very hard to use nanoelectronics in space just because of the hardening issue).

    It's fast enough, if you're using the right languages, since these probes don't have an OS to run- they only have commands.

    I could see the tech being custom-built, but you'd need a large heatsink to take that heat away (radiate it). But if you did that you could utilise the fact that processor design has come 20 years since these old processors were invented, and get something that runs quite a bit faster.
  • 0 Hide
    leon2006 , December 7, 2011 5:11 PM
    They need a radiation hardened version of it before it goes out in space
  • 0 Hide
    robochump , December 7, 2011 6:13 PM
    dark_knight33Ever seen "The Time Machine" (2002)? Not to mention the moon is mostly iron, so far as we know, and there is no shortage of iron on Earth, nor is it economically feasible to transport any material from the moon for commercial sale.



    In the distant future the Moon will become a launching site for Space exploration. As seen with the Apollo missions it takes much less effort to launch from the Moon than Earth! Sadly the economy will dictate when this is done and will most likely delay it far beyond what we are capable of doing if the funds were there.
  • 0 Hide
    kronos_cornelius , December 7, 2011 6:55 PM
    jamie_1318I don't think you can use Tesla in space. That CPU is way too sensitive to cosmic rays, like any other CPU designed to work in space. They use custom designed CPUs in anything that goes to space in order to avoid being totally fried by space-radiation. If anyone has any more information on the topic I'd be interested to hear.


    Yes, I remember reading that the current rover on Mars use technology that is at least 10 years old just because of the time it takes to adapt chip design to the requirements of space. I don't think they can use the latest nano-gate printing process because of what jamie_1318 mentions. I wonder if they are using shielding techniques to make this possible now.
  • 1 Hide
    Unolocogringo , December 7, 2011 7:21 PM
    Quote:
    Ever seen "The Time Machine" (2002)? Not to mention the moon is mostly iron

    According to a Discovery Channel series I was watching last night ,the moon has almost NO IRON.
    That is why scientist think the moon was created from cast off material whan a large asteroid or small planet collided with earth.
    The rocks there are of the same composition as rocks on the outer crust or mantle of the earth.
  • 0 Hide
    dalethepcman , December 7, 2011 8:35 PM
    LuckyDucky7I could see the tech being custom-built, but you'd need a large heatsink to take that heat away (radiate it)...


    How big a heat sink do you need for a location that has a temperature of -170f on the dark side? Any why would you land in the middle of the light side where temperatures reach 280f (above operating temperature of silicon chips)?
  • 0 Hide
    Unolocogringo , December 7, 2011 11:28 PM
    Chips would have to be designed to withstand both extremes.
    We only see one side of the moon, because it does not rotate on its axis like the earth. So we call it the dark side. But as the moon rotates arround the earth it has light and dark in all areas. Very similar to earth in that if you stood in one place you would have day and night times in relation to the sun.
  • 0 Hide
    hetneo , December 7, 2011 11:37 PM
    This is bound to fail, none of the consumer chips can withstand extreme conditions of space. Yes some CPUs worked well when cooled by liquid nitrogen or helium, but they performed well for few minutes, it's different thing to perform for days.
  • 0 Hide
    Unolocogringo , December 7, 2011 11:59 PM
    It is my understanding that the $*^ chips are used for space flight because of the 1000nm production process.

    With transistors this large,a stray electron (billions x trillions in space) hitting the chip would not cause an error. Tyhe smaller processes would cause errors when hit with stray electrons.

    It is not about heat,speed or energy consumption but exact calculations without errors.
    an error in the early calculations of flight path could send you to Mars or deep space and not back to earth.
  • 0 Hide
    Unolocogringo , December 8, 2011 12:03 AM
    I meant 486 above where you see $*^
Display more comments