Page 1 of 3 123 LastLast
Results 1 to 15 of 38
  1. #1

    Question waypoint averaging

    I was using a Magellan Triton 2000 to plot a fence line at a farm and frankly it bordered on useless. The terrain must be too hilly and has too much tree cover. Someone suggested that I leave the unit still at each important location to get an average location but it took me some time to realise that the Triton doesn't do that waypoint averaging, and indeed even that it was called "waypoint averaging" All I got was a trail that wandered tens of meters in all directions.

    Can anyone tell me how beneficial waypoint averaging is on a recreational gps unit such as a Garmin ? i.e. what sort of accuracy improvement does it provide ?

    Also, does the TwoNav Aventura support waypoint averaging ?

    cheers,
    - Alan

  2. #2
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    Hi and Alan,

    Not sure that I can answer the specific question regarding waypoint averaging, but I've commenced some testing.

    However, ... even with perfect reception, clear sky and flat ground, the best track accuracy you will achieve with any recreational GPS is +/- 5m.

    So say you walked a perfectly straight line, with 1 sec track capture, the best accuracy of any point captured would fall in the range +/- 5m. The points will most likely line up pretty nicely; if captured in the space of 15 minutes or so.

    Walk the same line, 24 hours later and you may get a different but similar straight'ish track.

    Average this over several captures and draw the median of the points and you'll be getting pretty close.

    It's the way OSM works. Many captures make for accurate tracks.
    Attached Images Attached Images
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  3. #3
    Join Date
    Feb 2006
    Location
    East Bentleigh Victoria Aust
    Posts
    3,057
    Can anyone tell me how beneficial waypoint averaging is on a recreational gps unit such as a Garmin ? i.e. what sort of accuracy improvement does it provide ?
    I use it on my Garmin 60Cx when standing still at a point I want to mark as accurately as possible, I wait till it has averaged about 50 recordings. It seems to help a lot.
    Also, does the TwoNav Aventura support waypoint averaging ?
    I don't think so, ie it just records a point on the assumed position the moment you press the 'Mark' button.
    regards Michael

    yacht racing at RMYS
    Lake Mountain Ski Patrol
    bush walking anywhere in Vic

  4. #4
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    Here is how it works on the latest Garmins.

    Improved accuracy by measuring key waypoints, with Garmin averaging, seems quite positive.
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  5. #5
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    Something to ponder when 'shooting for accuracy'.

    To attempt to obtain the highest accuracy; select the grid setting with the highest resolution (on your GPS). e.g.

    ddd mm' ss.s" resolves to approx 3.1m

    ddd mm.mmm' resolves to approx 1.86m

    ddd.ddddd resolves to approx 1.16m

    So ddd.ddddd ie degrees to 5 decimal places is probably as good a you'll get.

    As Garmin recommends, taking a number of averaged readings over time (and averaging them) will further improve the reading precision.

    For optimal results, at least 4-8 samples should be collected spaced at least 90 minutes apart (allowing the satellite constellation sufficient time to change).
    Attached Images Attached Images
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  6. #6
    Join Date
    Jun 2005
    Location
    Western Australia
    Posts
    3,294
    Quote Originally Posted by Wahroonga Farm
    Something to ponder when 'shooting for accuracy'.

    To attempt to obtain the highest accuracy; select the grid setting with the highest resolution (on your GPS). e.g.

    ddd mm' ss.s" resolves to approx 3.1m

    ddd mm.mmm' resolves to approx 1.86m

    ddd.ddddd resolves to approx 1.16m

    So ddd.ddddd ie degrees to 5 decimal places is probably as good a you'll get.

    Taking several averaged readings over time (and averaging them) will further improve the reading accuracy.

    ....and UTM? 1.0m?

    Cheers
    Greg
    To find yourself, think for yourself Socrates 469 BC

  7. #7
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    Quote Originally Posted by The Explorer
    ....and UTM? 1.0m?

    Cheers
    Greg
    By definition .
    Attached Images Attached Images
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  8. #8
    Join Date
    Feb 2006
    Location
    East Bentleigh Victoria Aust
    Posts
    3,057
    A distance of 5 metres on the ground is drawn on a 1:25000 map as a line 0.2 mm long, unless you have a very sharp pencil, you cannot plot anything to greater accuracy.
    After the 09 firesroared over Lake Mt we replaced all the ski trail marker posts. I recorded their positions by standing at each post & using averaging on the 60Cx.
    I am sure the positions I recorded are within 5m accuracy, & on the 1:25k map printed from OziExplorer, the ink dot is bigger than 0.2 mm.
    regards Michael

    yacht racing at RMYS
    Lake Mountain Ski Patrol
    bush walking anywhere in Vic

  9. #9
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    All depends which side of the 'pencil' your cow is grazing.
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  10. #10
    Join Date
    Feb 2006
    Location
    East Bentleigh Victoria Aust
    Posts
    3,057
    At Lake Mt we have wombats grazing.
    A 'herd' of wombats?
    Wonder what the collective noun is?
    regards Michael

    yacht racing at RMYS
    Lake Mountain Ski Patrol
    bush walking anywhere in Vic

  11. #11
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    Here are four averaged readings taken over about 8 hours on the trusty Oregon.

    There is no movement for the Easting so we can lock that in.

    There is some movement (9 to 1) for the Southing. This is a variation of 3 x 1.86 or 6m.

    However with averaging we're going to achieve a couple of metres.

    Not bad for a cheap and cheerful Garmin.

    .861 .484
    .859 .484
    .860 .484
    .861 .484

    Starting to lock it in. Go the Garmin.
    Attached Images Attached Images
    • File Type: jpg 4.jpg (157.5 KB, 2 views)
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  12. #12
    From what I read waypoint averaging can not work as many say.

  13. #13
    Join Date
    Aug 2007
    Location
    Gloucester, NSW
    Posts
    2,936
    Quote Originally Posted by Bug's
    From what I read waypoint averaging can not work as many say.
    What are the key problems?
    John k - Garmin Oregon 300, GPSmap76 & GPS16 plus various WinCe PNDs with OziCe & IGO8

  14. #14
    Join Date
    Jun 2005
    Location
    Western Australia
    Posts
    3,294
    Quote Originally Posted by Michael
    snip...A distance of 5 metres on the ground is drawn on a 1:25000 map as a line 0.2 mm long, unless you have a very sharp pencil, you cannot plot anything to greater accuracy.
    Unless of course you use a larger scale image (1:10,000, 1:5000 etc)....though getting an accurate position (in number form) and then being able to plot/visually display the point accurately are two different issues.

    Anyway - thought we were discussing accuracy of the different coordinate formats as read on the GPS. UTM appears to be the best to use to display/convey coordinates, as the number is to the nearest metre (averaged or not). Mining companies/Surveyors generally use metres (for this and other reasons). I have never seen coordinates written as degrees (in any format) on a peg in the bush that I can remember.


    Cheers
    Greg
    To find yourself, think for yourself Socrates 469 BC

  15. #15
    I thank you all for the helpful replies and for the welcome to the forum

    I've been using UTM on my triton 2000 to get the maximum precision but of course high precision does not ensure high accuracy.

    I'm convinced that for my needs waypoint averaging will be a significant advantage because the instantaneous location drift has been rather severe when I'm away from the hilltops. Then again, maybe I've got a defective receiver in my Triton.

    What I really want is 1-2m accuracy but toys with that capability are still rather expensive and also require subscription to a differential GPS data service. One day the wannabe surveyor in me will be satisfied.

    cheers
    - Alan

Page 1 of 3 123 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •