Forgot your password?
typodupeerror
Space Earth Technology

1.4 Billion Pixel Camera To Watch For Asteroids 138

Posted by timothy
from the one-last-hawaiian-vacation dept.
SpaceSlug writes "The world's largest digital camera is to be used to keep an eye out for asteroids heading towards Earth. The Panoramic Survey Telescope and Rapid Response System (Pan-STARRS) has been built by researchers at MIT's Lincoln Lab. At its heart is a 1.4 billion pixel (or 1400 megapixel) camera that will scan the night sky looking for rogue near-Earth objects from atop Mount Haleakala in Maui Island, Hawaii. The system uses something called an orthogonal transfer CCD to remove atmospheric blur from images."
This discussion has been archived. No new comments can be posted.

1.4 Billion Pixel Camera To Watch For Asteroids

Comments Filter:
  • Gigapixels is that???
  • by deft (253558) on Tuesday November 25, 2008 @01:44PM (#25888439) Homepage

    You just need to point her down at the beaches of Hawaii a few times a year and capture some of the scenery.

    can you say gigapixelboobs.com?

  • Pending Doom (Score:5, Interesting)

    by tripdizzle (1386273) on Tuesday November 25, 2008 @01:45PM (#25888477)
    So now we will be able to see asteroids that could slam into our planet and possibly end life, but then what? Hit it with a missile or go Armageddon style on it?
    • by Kagura (843695)
      Our response depends on how soon it'll impact our planet.
    • Re:Pending Doom (Score:4, Insightful)

      by Last_Available_Usern (756093) on Tuesday November 25, 2008 @01:58PM (#25888661)
      The problem with illiciting action is proving risk. No ones going to mitigate a one in a million possibility. However, if you find some near-Earth objects that you can show have a 25% chance of hitting the Earth in the next 50 years, you might see a lot more development in the way of mitigation (or disaster planning at least).
    • Re:Pending Doom (Score:5, Insightful)

      by evanbd (210358) on Tuesday November 25, 2008 @02:18PM (#25888983)

      For starters, it's worth worrying about asteroids that would merely destroy a city rather than end life as we know it. And, if you spot them early, there are a number of techniques that could deflect them. With plenty of time to work, small changes in velocity can cause large changes in position years in the future -- turning an impact into a near miss. This is especially true if there is a close approach to another body before the impact, as small changes in position at the approach turn into larger changes in velocity.

      If you only need a tiny course correction, there are plenty of options. A gravitational tug, for example (put a spacecraft near the asteroid, use ion engines to maintain position, and let gravitational forces pull the asteroid toward the ship, and vice versa). That lets you use an ion engine to nudge the asteroid without solving the problems of landing on it or grabbing it. If you can get away with even less total impulse, you can simply paint a large portion of it white and let light pressure from the Sun do the work for you.

      Things like large rocket engines and nuclear blasts are crude, blunt instruments; if you have warning, a more subtle approach is appropriate.

      • Re:Pending Doom (Score:5, Insightful)

        by HarvardAce (771954) on Tuesday November 25, 2008 @02:43PM (#25889319) Homepage

        With plenty of time to work, small changes in velocity can cause large changes in position years in the future -- turning an impact into a near miss.

        Or, given the fact that even the most advanced prediction algorithms still have to cut some corners (therefore leading to some uncertainty), it could turn a near miss into an impact.

        • Re: (Score:2, Interesting)

          by Tablizer (95088)

          Or, given the fact that even the most advanced prediction algorithms still have to cut some corners (therefore leading to some uncertainty)

          Cut corners? You mean along the lines of using type Float instead of Double? More likely the errors are due to natural issues such as sun-light reflecting off the surface of the roid in unknown ways giving it a slight push or imprecise knowledge about Jupiter's gravity profile at given distances. When a roid passes close to a planet, small differences in path can be grea

          • More likely the errors are due to natural issues such as sun-light reflecting off the surface of the roid in unknown ways giving it a slight push or imprecise knowledge about Jupiter's gravity profile at given distances.

            This is exactly what I meant by "cut corners." Probably not the best way to describe it, but there are, at the roughest level, three different items that lead to error in predicting trajectories. The first is errors/inprecision in measurements, which this camera will help reduce. The second is us not knowing exactly how certain forces will affect a particular asteroid. The last is the inability to accurately simulate the forces that we do know, due to computational complexity. It is here where things a

        • At issue with many of our current NEO detections is that it takes astronomers time to take enough measurements to calculate an accurate orbit. PanSTARRS will be taking those measurements nightly, or at worst every other day, which will speed the process of drawing the curves considerably.

          • Accurately predicting an asteroid's orbit is much more than simply knowing the current position and the current velocity. Even if you knew them, exactly, there are many more variables that affect the trajectory of an asteroid. Current prediction models still have trouble with simple things like gravity (most models only take into account the sun, planets, satellites of the planets, and a few of the larger astroids/minor planets, and all but the most advanced models treat them as point masses which they ob
        • by robbak (775424)

          This device is all about finding them. Once you have found something that has an orbit that brings it close to the earth, hundreds of telescopes will be pointed at it, leading to many highly accurate measurements that will specify its location and orbit to a nice precision.
          This device will find asteroids with a 1 on 1000 chance of impacting. Withing a few days, we'll know to 1 in 100. Give it a week, month or year, and we'll know what street to paint an X on.

          • The point I was trying to make is that there are a lot of forces acting on an asteroid while it is in orbit. As others have said, there are things like spin, close encounters with other unknown asteroids, the exact gravity profile of the planets, etc. Location and current velocity are just two of the many variables that affect the trajectory of an asteroid. Current projections can estimate the amount of error that each of these factors can cause, but they cannot, with today's technology, accurately accou
            • All this means is that your trajectory change ought to be done with serious overkill so as to definitively push it outside of the entire error basket. If caught years in advance this will still men an absolutely miniscule impulse.

    • by east coast (590680) on Tuesday November 25, 2008 @02:44PM (#25889327)
      Hit it with a missile or go Armageddon style on it?

      As long as it gets rid of Bruce Willis it's a win-win situation.
      • As long as it gets rid of Bruce Willis it's a win-win situation. And keeps Liv Tyler, and I become her for a day, and go jogging all day holding a mirror in front of me.
        • by dkleinsc (563838)

          Liv Tyler? You must be new here: the celebrity that is kept has to be Natalie Portman (petrified in hot grits).

    • by Falkkin (97268)

      The book "Death from the Skies" by Phil Plait has a lengthy chapter devoted specifically to asteroid impacts and how we might consider avoiding them. The missile idea is a bad one, as the individual fragments will still hit Earth and quite possibly do more damage than the original asteroid due to immediately affecting an even wider area. A couple more promising ideas are gravitational deflection (park some other massive object nearby and allow the new object's gravity to slowly move the asteroid off a col

      • by nasor (690345)
        The usual point of the "missile idea" isn't to destroy the asteroid, but merely to alter its course. A 1 megaton bomb detonated 1 km from an asteroid will impart about 300 megajoules/square meter of asteroid. That should be more than enough to vaporize a layer along the outer surface, which will dramatically later its course. In terms of simplicity, nothing competes with nuking the asteroid. A gravity tug is clever and all, but it assumes that you can match velocities with the asteroid and then maintain you
        • by robbak (775424)

          Um, no. You have been watching too many movies.

          An explosion in space does - practically nothing. An explosion on earth pushes the air away in a big shock wave that does the damage. An explosion in space dissipates immediately with little effect except to the item that exploded. (related phenomenon: Why the lunar landers produced so little dust, and no crater.)
          For an explosive device to do much, it would have to be embedded into the asteroid and blow a chunk off the side. Then conservation of momentum would

          • Re: (Score:3, Informative)

            Um,no. You have missed the point.

            It has nothing to do with a shock wave. The theory is that the detonation will heat the surface of the asteroid causing a thin layer to vaporize and move away from the body thus imparting a small thrust and altering the trajectory.

          • by nasor (690345)
            Re-read my post. The "shockwave" doesn't have anything to do with it. Nuclear bombs throw out a HUGE amount of energy in the form of xrays and hard UV, easily enough to vaporize the outer surface of the asteroid and radically change its course. I even included some exact numbers on how much energy would be deposited on the asteroid based on a certain bomb size detonating at a certain distance, which of course you ignored.

            A gravity tug is MUCH more complicated than simply nuking the asteroid, because it r
            • by robbak (775424)

              Thank you: You are, of course, completely correct. I do not know how I managed to so badly mis-read your comment.

              Again, sorry and thank you.

    • by Kleen13 (1006327)
      Both. Good motivation for the drill team. Just don't tell them why their uniforms are redshirts. http://en.wikipedia.org/wiki/Redshirt_(character) [wikipedia.org]
    • So now we will be able to see asteroids that could slam into our planet and possibly end life

      No, we will be able to see asteroids that could slam into our planet and possibly end life in the northern hemisphere. And then we can send a bomb up or something to move it away. Who cares about the southern hemisphere. :D

  • Damn (Score:3, Funny)

    by MightyMartian (840721) on Tuesday November 25, 2008 @01:46PM (#25888493) Journal

    Damn, now it means that when my brother-in-law sets his next camera to maximum resolution (as he always does), I'm going to get 50gb image files.

  • by zappepcs (820751) on Tuesday November 25, 2008 @01:51PM (#25888569) Journal

    Last time I looked, Asteroids was at the local bowling alley. Do I win a prize? Do these youngster space explorer types need any more investigative help? All the comets they need are under my kitchen sink! there! ba da bing! Oh, my neighbor guy has Saturn in his driveway! But if you guys need help or a camera to find Uranus... damn, I'm out!

  • Blah... (Score:1, Offtopic)

    Wake me up when Apple puts it in the next iPhone.
  • Kodak moment (Score:2, Insightful)

    Well, it's almost as good as what the NSA uses to spy on you with. Aren't you glad we have our priorities straight in this country?

    • by SlashSnot (647926)
      Pretty sure your priorites are straight. As is your tin foil hat across your brow.
  • Not sure if there is much "blue shift" that can be detected from a near-earth object, such as an asteroid. So, maybe we should put two up there to add a little perspective, sensu Grover.
    • by boris111 (837756)

      I'm no astronomer, but I believe you can triangulate objects based on the earth's position around the sun.

      • by wjh31 (1372867)
        yes, but to do that you have to wait for the earth to move to the other side first
        • Re: (Score:2, Informative)

          by JayAitch (1277640)

          Maybe for very far objects outside the solar system, but for these objects the earth moves a significant distance within a day. Backyard astronomers do this all the time.

    • by Darth_brooks (180756) <clipper377@gm3.1415926ail.com minus pi> on Tuesday November 25, 2008 @02:04PM (#25888769) Homepage

      Asteroid hunting doesn't really have anything to do with blue or red shifting. You're not looking to see whether a distant object is moving towards or away form you. More likely, they're looking at dots. Specifically, which dots in picture A moved in comparison to picture B and which one didn't.

      Think of it this way: Step out at night and look at the stars and whatever planet happens to be in view. Now, step out the next night at precisely the same time (ok, to be fair, a couple minutes later) and look again. The stars are in the same spot, but the planet has moved.

      With high-res digital cameras you can take very precise pictures, then let software pick out which of the faint dots are distant stars, and which maybe be asteroids. It's a pretty standard way of discovering and plotting the course of the various odds and ends floating around our solar system.

      • Upon further reflection, and with your succinct description, I think I understand. Obviously, stereoscopic view isn't needed to track orbits, most telescopes are monocular. One just needs to see something moving (e.g., a tool-bag in orbit [spaceweather.com]). I guess for some reason the impression of an object heading straight at the lens came to mind, which is clearly, well silly.
  • So just how many near Earth objects come from atop Haleakala?
  • by Anonymous Coward

    I gots to know

  • No, it's a lost toolbag! So many practical applications for things lost in space.
    • LOL, i could have imagined her screaming, "Oh noes, a $100,000 bag of tools is floating away from me at a snail's pace, wtf do I do!!!!!!!"
  • Astronomy Magazine (Score:3, Interesting)

    by Neuropol (665537) * on Tuesday November 25, 2008 @02:03PM (#25888747) Homepage
    I think i just read some thing about this in Astronomy Magazine from earlier this year. The sensitivity and rapid ability to shoot large areas in a short amount of time will allow for this telescope to scan and record the entire Hawaiian skies every 3 days in search of Asteroids, Supernovae, and other phenomenon.
  • Blurred summary (Score:5, Informative)

    by mcgrew (92797) * on Tuesday November 25, 2008 @02:11PM (#25888867) Homepage Journal

    The system uses something called an orthogonal transfer CCD to remove atmospheric blur from images.

    Shoddy. "Something called?" Come on, guys, this is supposed to be "news for nerds". If you can't find it on wikipedia, use google.

    orthogonal transfer CCD (OTCCD) [harvard.edu]

    We have designed and built a new type of CCD that we call an orthogonal transfer CCD (OTCCD), which permits parallel clocking horizontally as well as vertically. The device has been used successfully to remove image motion caused by atmospheric turbulence at rates up to 100 Hz, and promises to be a better, cheaper way to carry out image motion correction for imaging than by using fast tip/tilt mirrors. We report on the device characteristics, and find that the large number of transfers needed to track image motion does not significantly degrade the image either because of charge transfer inefficiency or because of charge traps. For example, after 100 sec of tracking at 100 Hz approximately 3% of the charge would diffuse into a skirt around the point spread function. Four nights of data at the Michigan-Dartmouth-MIT (MDM) 2.4-m telescope also indicate that the atmosphere is surprisingly benign, in terms of both the speed and coherence angle of image motion. Image motion compensation improved image sharpness by about 0.5'' in quadrature with no degradation over a field of at least 3 arcminutes. (SECTION: Astronomical Instrumentation)

    • Reading this my question would be why they don't use the movement technology used in optical image stabilization for digital cameras (video and still) which has been around for quite a long time... some techniques move a lens, which is the last element before the sensor, up and down while others move the sensor up and down (which still counts as OIS it seems). I would assume that in order to achieve its goal OIS is capable of movements accurate to a pixel width or less but perhaps it isn't that good? Or are
      • No mechanical parts. It's probably more reliable, and surely needs less maintenance.
        Also, it might be tricky to move a 40cm x 40cm detector array around at 100Hz without
        introducing jitter or tilt.
  • by Channard (693317) on Tuesday November 25, 2008 @02:11PM (#25888881) Journal
    ... the manufacturers only included a 16mb SD card.
  • by rpiquepa (644694) on Tuesday November 25, 2008 @02:12PM (#25888883) Homepage
    You also should read a story published 4 days ago about this camera by ZDNet. Here is a link [zdnet.com] to this article, which contains several pictures not included in the Technology Review article.
  • pieces of ass, umm, fully-feeling, umm, screen-fulfilling asTEROIDS on display at one time... Talk about "getting your rocks off"....

  • Watch them still miss the asteroid because it was never detected due to a dead pixel. If Hollywood has taught me anything, the impossible will always happen. I'm not worried though, because it also taught me that the problem will only last a few hours and everything will be solved in the last few minutes.
  • What kind of lenses? (Score:2, Interesting)

    by qwertphobia (825473)

    I bet it has no practical depth of field, even when stopped down to f/64. Wonder what the pixel density is...

    • It is an interesting question, though I'm not going to rtfa to see if they have an overall sensor size. Luckily, when imaging most celestial opbjects the DOF should be pretty insignificant. On the bright side, if an asteroid does come out blurry due to a depth of field issue, we're probably not going to be around long enough to worry about it!

    • There is absolutely no need to stop down a lens like this- everything is infinity focused.

  • I don't know about anyone else but I find stuff like this to be really inspiring. Why is that astronomy in general doesn't get more mainstream media coverage?

    • by edremy (36408)
      You want my take? Light pollution. 100+ years ago you could look up at the sky at night and actually see the stars. Today? Bah- you're lucky if you live in an area that you can see magnitude 3-4 stars where a decent dark adapted eye can see down to 6 easily, and the difference is stunning. City dwellers have it even worse.

      Go camping in Yosemite or the high western desert some day and backpack away from other people. Let your eyes adapt at night for a while and then look up. Compare that to what you

  • The system uses something called an orthogonal transfer CCD to remove atmospheric blur from images.

    Hopefully they've performed some real-world testing to ensure this technology doesn't also remove, you know, ASTEROIDS.

  • You know.. (Score:1, Redundant)

    by SnarfQuest (469614)

    They really just want this camera for porn.

  • ...if an asteroid is big enough to end life as we know it on earth, are 1.4 gigapixels really necessary?
  • ...see a deep-core drilling team on the asteroid's surface?
  • Imagine that you somehow get to stand on Mars let's say...
    What do you see? Well, we don't really know do we? Because ALL OF THE DAMN CAMERAS that Nasa or whomever send up there in the firmament, to the heavenly bodies or the moons are not designed for human vision.
    No. The imaging data sent back to terrafirma has to be 'processed' with 'algorithms' before we can see a representation of human vision!
    This 'representation' is not accurate, as captions often state that the image in question is made up of various

  • They just don't have one of these cameras, they have four. Each pixel will be 2 bytes therefore 11.2GB per shot. Exposure times will be 30 seconds, so that is 1.3TB of data generated per hour.

    That is going to be some serious number crunching when you need to compare at least two images. You want to finish all that comparison work (possible alignment work as well) before the following night (14hrs?). The data is going to be stored with Microsoft SQL Server. Storing and retrieving images sure, but when i

    • by Shag (3737)

      They just don't have one of these cameras, they have four.

      Uh, no. They have one telescope with one camera on it. The system on Haleakala is the prototype, called PS-1.

      The original plan was to build a final system with 4 scopes and 4 cameras, PS-4. Probably on Mauna Kea. Probably where the UH 2.2-meter (which I operate) is now. But they have to go through permitting and everything.

      • by kramulous (977841) *

        You lucky bastard.

        Does your role cover the hardware maintenance/upkeep (telescope, compute machines, storage), or the software implementation/number crunching? Or both?

        • by Shag (3737)

          Hardware and software upkeep? Uh, a tiny bit, but really mostly dealing with wetware. I'm operating the telescope, and as a former sysadmin, rather prefer to not have root on the boxen (sudo, where needed, will suffice). Server users usually don't (although they should!) question whether your racks are properly aligned with the stars.

          • by kramulous (977841) *

            Hey,

            By strange coincidence the guys at the local planetarium [wikipedia.org] have said that I can feel free to use their skydome system (within reason, but I think they are also interested in the visualisation work I do). The trade off is that I have to help the local Astronomy club get their datasets into the existing software - fair trade. But it only handles basic geometry, so I want to port some of my codes to the projection system. The parallel rendering (8 machines, each a single projector) will be

APL is a write-only language. I can write programs in APL, but I can't read any of them. -- Roy Keir

Working...