Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Upgrades Technology Science

A New Technique Makes GPS Accurate To An Inch (gizmodo.com) 127

A team from the University of California, Riverside, has developed a technique that augments regular GPS data with on-board inertial measurements from a sensor. Actually, that's been tried before, but in the past it's required large computers to combine the two data streams, rendering it ineffective for use in cars or mobile devices. Instead what the University of California team has done is create a set of new algorithms which, it claims, reduce the complexity of the calculation by several orders of magnitude. In turn, that allows GPS systems in a mobile device to calculate position with an accuracy of just an inch.
This discussion has been archived. No new comments can be posted.

A New Technique Makes GPS Accurate To An Inch

Comments Filter:
  • by mikael ( 484 ) on Saturday February 13, 2016 @02:53PM (#51501603)

    One of the most time consuming parts of a short journey is getting the passenger doors aligned with the port-side gangways. Unlike airports, it's not the gangways that move to the plane, it's the vessel that must align with the portside. Sometimes the portside gangway can move up or down, but many times, the crew have to tie down these mini gangways with ropes when the tides and ballast tanks aren't enough. It takes several minutes of maneuvering to get the ship aligned with the dockside, sometimes even having to reverse and try again, especially in heavy swells. If they could get GPS down to several inches, combined with the sideways movement that many catamarans have, docking could be done automatically.

    • by Calydor ( 739835 ) on Saturday February 13, 2016 @03:03PM (#51501663)

      Couldn't the same level of 'automatic' be achieved with image recognition cameras at the doors, or other sensors to achieve the same result?

      • by zm ( 257549 ) on Saturday February 13, 2016 @03:07PM (#51501697) Homepage

        Couldn't the same level of 'automatic' be achieved with image recognition cameras at the doors, or other sensors to achieve the same result?

        Why would you want to do it in a simple way?

      • Why try to maneuver the huge vessel? A simple floating bridge with gangways manually controlled like they do with airbridges for the airplanes is even more simple. Given the mass and inertia of the ships, throwing a couple of thick ropes and tightening them automatically will adjust the floating bridge gangway to the ship.
      • Scanning Lidar will work. Map it first and run some ICP algorithms. Basically you have a GPS pose and a ICP Lidar corrected pose. Highly accurate if you can map the dock well enough.

        • by ShaunC ( 203807 )

          I never knew Juggalos were so concerned with positioning.

          • by KGIII ( 973947 )

            I'm still convinced the 'net could use a "smack user" button. That was horrible and you should be ashamed of yourself! Why, I ought to...

            I have a niece who's into that thing. She even wears war paint and goes out to festivals and whatnot. I think they call them gatherings. I've met some of them and I've been given the title of "Honorary Juggalo." That's not exactly something I was looking forward to gaining as an accomplishment but, well... It's something and I guess it's good that they have each other.

            The

    • One of the most time consuming parts of a short journey is getting the passenger doors aligned with the port-side gangways. Unlike airports, it's not the gangways that move to the plane, it's the vessel that must align with the portside. Sometimes the portside gangway can move up or down, but many times, the crew have to tie down these mini gangways with ropes when the tides and ballast tanks aren't enough. It takes several minutes of maneuvering to get the ship aligned with the dockside, sometimes even having to reverse and try again, especially in heavy swells. If they could get GPS down to several inches, combined with the sideways movement that many catamarans have, docking could be done automatically.

      Maybe they'd have better luck with the starboard side.

    • If they could get GPS down to several inches, combined with the sideways movement that many catamarans have, docking could be done automatically.

      That would shift the bottleneck to engine power, and the deliverability of that power. That costs serious coin for any working vessel,

  • Encrypted (Score:3, Interesting)

    by Anonymous Coward on Saturday February 13, 2016 @02:54PM (#51501611)

    Given that we can now use GPS to great accuracy. Does this mean that the US military no longer needs to encrypt the end of the GPS signal? After all, the military has always been able to use GPS for very precise location, whereas civilians had to put up with very coarse location.

    • Re:Encrypted (Score:5, Informative)

      by Anonymous Coward on Saturday February 13, 2016 @03:29PM (#51501813)

      This is incorrect. President Clinton removed selective availability from GPS. That is why we have the location boom today. We are at as accurate a position as we can be with the GPS reception. This happened in 2000. There are other forms of signal degradation that are available. SA is not an issue anymore. I am a professional land surveyor. Multi-pathing is currently your greatest hurdle to overcoming highly accurate positioning. That unfortunately, requires fairly complex calculations to remove it properly. Even with my highly accurate receivers, I have to remove multi-pathing manually sometimes. Reflection is a harsh mistress.

      • Re: (Score:3, Informative)

        by DerekLyons ( 302214 )

        I am a professional land surveyor.

        You're a moron with no clue what you're talking about.

        President Clinton removed selective availability from GPS. That is why we have the location boom today. We are at as accurate a position as we can be with the GPS reception.

        President Clinton turned off Selective Availability on the C/A (coarse acquisition) signal. The more accurate P(Y) (precision location) signal used by the military is still encrypted [wikipedia.org].

    • Like others have said, there is no Civilian degradation on purpose anymore. That was the SA system, that is off for several years and is not coming back at least because it is no longer a effective defense against foreign military parties. The only thing the Encrypted GPS code provides is AS (anti-spoofing). If used this avoids things like Iran taking control of your baby drone.
      • Re:Encrypted (Score:4, Informative)

        by russotto ( 537200 ) on Saturday February 13, 2016 @04:55PM (#51502339) Journal

        The P(Y) code offers a couple of advantages besides anti-spoof

        1) Faster code rate for more precise positioning, also offered by the newer civilian signals (L1C, L2C, and L5)

        2) It exists on both L1 and L2, allowing the receiver to more accurately model the atmospheric delay terms, reducing that source of error. This is also provided by the L2C and L5 signals, but not all satellites yet transmit them.

  • I'm sure there will be a lot of obvious applications for this technology, but I can see robotics being a big one.

    GPS hasn't been practical for robotics but with this level of accuracy, I wouldn't be surprised to many robotic applications currently being done by humans.

    It would be interesting to see how the algorithm keeps its accuracy over time and distance.

    • by skids ( 119237 )

      If the new algorithm also helps indoor positioning enough to be accurate over a decent sized floorplan, it could be useful for indoor wifi surveys when deploying or upgrading enterprise networks. Last time I looked there still was not a product that didn't require manual entry of your position on a map during a survey, which is error prone and drives up manpower costs. Eventually WiFi clients could be extended to report back position information to the network so dead spots could be eliminated without a s

  • The rest of the world gets it "accurate to a centimetre".
  • by Anonymous Coward

    ...nobody actually uses GPS for anything life-critical requiring accuracy, because signals can be jammed/spoofed, and sometimes the calculation is off due to reflections or satellites behaving not as expected.

    And by "nobody", I except the military, but nobody notices when they miss their target.

    • Comment removed based on user account deletion
    • Then there is every time an airplane uses GPS to find where the airport and runway are, i.e.an RNAV approach.

      ...laura

  • by Anonymous Coward

    http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7349142

  • Accurate GPS is good.

    But I'd really much prefer a GPS that can work indoors, in cities with tall buildings, near hills and mountains etc.

    That seems to have much more uses than getting something from a handful of inches down to fractions of an inch.

    My car and phone sometimes get confused about precisely where I am and which turn-off I've taken. And in Belgium (where there are a LOT of underground roads), it barely works at all - by the time it locks on, I've had to go down another tunnel. In Central London

    • Re:Signal (Score:5, Funny)

      by Saeed al-Sahaf ( 665390 ) on Saturday February 13, 2016 @03:24PM (#51501791) Homepage
      Yes, I know what you mean. I left my home in upstate NY to buy a chalupa at the Taco Bell down the street and ended up in Nova Scotia...
      • Don't you look at the road, and check for border crossings? Particularly an international border crossing that you'd have to pass to end up in Nova Scotia?
    • Regarding indoor use: In my home I have erected walls around each room with well defined travel ways identifiable both by visual and tactile means to transition from one room to another. In addition I have labeled each of the designated sleep areas with the appropriate occupant identity. To date this system has proven adequate.

    • Actually, more than cities w/ tall buildings, I'd like it if GPS could - if one is in a multi-level parking lot, say at a mall, detect which level the car is in. Of course, that's unlikely, given the penetration needed within.
  • I wonder how the Three Letter Agencies will respond to this? Military GPS has always had access to more accurate coordinates, now anyone can have it? Someone in Maryland is shitting bricks.
  • Now you won't have to hire a surveyor to find out where the edge of your property is.
    • Now you won't have to hire a surveyor to find out where the edge of your property is.

      That would require your property boundaries to be described in coordinates, rather than beginning at the iron pin 20 feet east south east from the largest oak tree; thence northwest 1 furlong until the stone wall bounding farmer Jones' land. More accurate GPS isn't really going to help you interpret something like this [bid4assets.com]. At least the US doesn't have to deal with issues like changes to the length of a furlong [wikipedia.org] made by Queen El

      • by Teun ( 17872 )
        Yeah right, the US has it's own odd measurement, the US survey mile:
        https://en.wikipedia.org/wiki/Mile#US_survey_mile

        Speaking about odd, what about feet, inches and yards?
        ;)
        • No, sadly it's not even that nice. The State Plane Coordinate System uses "feet" and by feet I mean either the "International Foot" which is defined as .3048 meters or the "U.S. Survey Foot" which is 1200/3937 meters. Sadly both are still in use today.

          http://vterrain.org/Projection... [vterrain.org]

      • How would you deal with continental drift? North America and Eurasia are moving apart by 2.5 cm per year.

        At the very least this could be used to tare or zero the readings from a local survey monument.

        Whats funny is the inch is officially defined as 25.4 mm. Dual scale measuring tape confirms this at multiples, as does using a unit converter [google.com]

  • When we reference "GPS", we primarily reference the US system.

    Galileo is a European civilian system, and once all 30 satellites are in orbit it will be centimeter accurate by default, with no accuracy degradation at the whim of US military.

    The bad news is the US is not happy. And had threatened jamming and shooting down the satellites in time of conflict. For the US this is perpetual or course. Should be interesting as the system is scheduled to be fully operational by 2020.
  • by fahrbot-bot ( 874524 ) on Saturday February 13, 2016 @03:32PM (#51501821)

    In turn, that allows GPS systems in a mobile device to calculate position with an accuracy of just an inch.

    Dick measuring goes high-tech.

  • So can I blame Jamie Condliffe for taking an IEEE article in metric and converting to imperial?
    After all, Gizmodo is a tech lite site; so you think they would want to culturally lead the way in dropping a unit of measure that no other country uses anymore.

  • by Mirar ( 264502 )

    The accelerometers on the phone are usually not that good, at all. "A mobile device" yes, but that's nothing new - the sensor fusion technology is old (kalman).
    Anyone that figured out what the new part is?

  • From a technical standpoint this is very cool. However, from a privacy/security standpoint, I am legitimately concerned that government could really misuse and abuse something like this.
  • by Garion911 ( 10618 ) on Saturday February 13, 2016 @04:52PM (#51502321) Homepage

    I used to work at a precision GPS company, on differential GPS. What they would do is have a 'base station' that would stay in a single spot, and average the position from the GPS signal for a period of time. This is because, due to atmospheric interference, the position 'wobbles'.. Once we have an average position, we use that position to come up with correctors to send to the mobile units (via radio modem usually, though other means are possible). This got us to be on par with what this article is claiming.

    I wonder if they account for the 'position wobble' of atmospheric interference. I suspect its possible, as they just pick one position they've received, and use the inertial adjust for the correctors. Not much more work than we were doing.

    (I didnt write any of the algorithms or anything, just shuffled data around to different devices and libraries.)

    • Ahh, finally you have shown me what to do with my antique Garmin GPS-12s. They have serial ports, and I have ESP-01s and level shifters lying around. DGPS, here I come

  • I believe that ship based dead reckoning systems are physically attached to the ship and "up" for the dead reckoning system is always "up" with respect the the physical ship. Same thing for "forward" or "front" (or "fore" if you want to get all nautical). How could this work for a phone that is in some random (and changing) orientation in your pocket?
    • Recent models can measure rotations as well as accelerations. The gyroscopes do drift a little bit, but the orientation vector is kept aligned in the long term using average acceleration (gravity), GPS position changes (horizontal speed vector direction), and the built-in compass (not very accurate, but it helps).

      Anyway, I doubt these researchers were using an iPhone, they probably had much better (and more expensive) sensors.

  • I use a usb gps sensor I built around 1998 which simply measured rate of change, connected via CAN to the speedometer and power steering, considered angle travel as well as using a compass/gimble. Certainly, it wasn't real precise, but it worked quite well. I don't think I used more than two hours in matlab to code it.

    As for CPU, it's a 16mhz z80 derivative... no FP.
    • So.. You coded something 20 years so in Matlab that runs on a z80?

      Umm. No. Stop making shit up you are just embarrassing yourself.

      Yes. Dead reckoning systems are not complicated.. But have many many issues..

      But you cannot run MATLAB on a z80.

      • by Khyber ( 864651 )

        But MATLAB has a Z80 simulator, you fecking moron, but if you actually used MATLAB, you wouldn't have made that stupid remark you just made.

  • A product with these specifications seems to be already available on the market: MTi-G [xsens.com] by Xsens. The technical specifications talk of a resolution of 2.5 cm, which is about an inch. It uses the kind of sensor fusion algorithms described in the article. Xsens is a Fairchild Semiconductor company, an industry icon delivering power solutions for the mobile, industrial, cloud, automotive, lighting, and computing industries. Xsens has offices in Enschede, the Netherlands and Los Angeles, California.
  • Yawn, come back when it's 1cm, then I'm impressed.

If all else fails, lower your standards.

Working...