Shop OBEX P1 Docs P2 Docs Learn Events
GPS question about similarity between units — Parallax Forums

GPS question about similarity between units

W9GFOW9GFO Posts: 4,010
edited 2010-03-30 06:31 in Accessories
If I don't get an answer to this I will run a test with what I have - probably will anyway...

Will two GPS modules, separated by just a few inches, show the same coordinates? Or at least track with each other in their deviation?

I know that the precise location reported will vary, but will the two units vary consistently by the same amount?

Would it make a difference if the two modules are not identical?

Rich H

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Simple Servo Tester, a kit from Gadget Gangster.

Comments

  • kf4ixmkf4ixm Posts: 529
    edited 2010-03-26 20:55
    I haven't tested that either, but i would suspect that they would show pretty much the same coordinates. i doubt being two different modules would make much difference, one may be able to track more sats than the other, or the antenna may have more gain than the other, making it somewhat more accurate. Good question though, got me to thinking...
  • John AbshierJohn Abshier Posts: 1,116
    edited 2010-03-26 22:07
    I doubt it, but would be interested in your results. I just logged 3 minutes with WAAS on a Parallax PMB-688. GPS was stationary. It took a 10x10 meter box to contain the readings. The CEP was 3.8 meters. Yesterday with no WAAS lock the box was 4.5x6.5 meters

    John Abshier
  • W9GFOW9GFO Posts: 4,010
    edited 2010-03-26 22:21
    What I am interested in is wether two units will have the same error, will they both show the same location within "the box"? Will they dance around inside the box as a couple or as individuals?

    Rich H

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    The Simple Servo Tester, a kit from Gadget Gangster.
  • FranklinFranklin Posts: 4,747
    edited 2010-03-26 22:37
    Interesting. i.e. is the error caused by the receivers or the senders?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    - Stephen
  • max72max72 Posts: 1,155
    edited 2010-03-27 14:27
    I haven't tested it, and it is an interesting question.
    This is my guess:

    In theory the main error source is related to atmosphere. This is why with a differential GPS you can achieve a 1-2 cm precision (both units are affected by the same atmospheric induced errors, so they are excluded). To do that the units must exchange informations, and there is a lot of processing involved.

    2 units with the same error source but operating independently will probably have different results. If they are using different firmware it will be more evident. With two identical units it might depend on the recent history, so you'll probably see them dancing togheter only for a fraction of the time.

    Consider with my DGPS changing firmware version makes the fixing different, and you can feel it.

    Massimo
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2010-03-29 19:02
    Actually, some of the entries into RoboMagellan and many/most of the entires into the DARPA challenge use multiple GPS receivers. I believe the results are averaged if within a certain margin of error. Outside that margin and a failure is usually assumed. But you can't afford to have a single GPS stop working in something driving a full-sized vehicle, so they do rely on more than one GPS source. They should report the same coordinates if the same chipset. Bear in mind that, of the units we carry, there are two different chipsets.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage

    Parallax Engineering
    ·
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-03-29 20:42
    As long as two identical GPS receivers are using the same set of satellites, they should track each other. (This is something that I, too, intend to test.) It's during the transition periods as satellites leave or come into range that the biggest differences will occur, since the units can lose or acquire the sats at different moments.

    -Phil
  • sylvie369sylvie369 Posts: 1,622
    edited 2010-03-29 21:45
    On Saturday I carried a Byonics GPS2 transmitting through an XBee to my laptop while also carrying a Garmin Vista HCX. Attached is an image of the two tracks. The divergences in the lower left corner are due to the XBee losing connection, but all of the stuff in the right half of the image should be good data from both units. For an idea of the scale, the yellow line is .27 miles long.
    GarminblueXBeered.JPG
    926 x 737 - 89K
  • W9GFOW9GFO Posts: 4,010
    edited 2010-03-29 23:01
    I have been testing two Polstar GPS units for the last hour or so. I am using the GPS Float Lite Demo from the Obex. Unfortunately it only displays in decimal degrees - to five places in latitude and four places in longitude. That is a granularity of 4ft in latitude and 24ft in longitude. Is there a object that is more precise? Or is that the limit of what the NMEA data is?

    As far as the two units tracking each other, they do. They often change at the same moment, sometimes one will lag about a second behind the other.

    I'd really like to see the data out to a couple more decimal places though.

    Rich H

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    The Simple Servo Tester, a kit from Gadget Gangster.
  • max72max72 Posts: 1,155
    edited 2010-03-30 06:31
    Check this one.

    obex.parallax.com/objects/225/

    It parses the string as they are. You must convert them in case.

    Massimo
Sign In or Register to comment.