| Author |
 Topic  |
|
|
john parrott
United Kingdom
60 Posts |
 Posted - 20/08/2011 : 18:46:03
|
This is more a question about GPS than MM, but hopefully there may be MM users who have investigated this.
I have 3 Garmin GPSmap60CSx hand-held GPS units. With reasonable satellite coverage, they are usually accurate to within 4 - 5 metres.
Can I improve on this, by using one (or both) of the units as a stationary 'base station' ? and if so, how? |
|
|
Irvine
United Kingdom
51 Posts |
Posted - 23/08/2011 : 09:41:00
|
John
The last time I used differential correction(20 years ago)the base station was a dedicated unit attached to a desktop computer. It logged the GPS position against its known position several times a minute and calculated and logged the error.
When we got back to base we had to download our waypoints into the desktop and the software made the appropriate corrections according to the recorded time.
I would guess you could use one of your Garmins as the base receiver but finding the software to log and apply the correction to your waypoints would be the tricky bit.
Good luck
Irvine |
 |
|
|
mapmaker
470 Posts |
Posted - 23/08/2011 : 10:25:43
|
| We experimented with this a couple of years ago with a view to building in functions to MM to facilitate this. We placed two identical gps receivers next to each other and monitored the "position" on both. Distressingly, not only did these stationary receivers wander by several metres they wandered differently which made us conclude that the idea would not work. |
|
|
|
djprang
36 Posts |
Posted - 23/08/2011 : 15:18:51
|
| MM - How sad to see that you gave up on this after one experiment such a long time ago. |
|
|
|
djprang
36 Posts |
Posted - 23/08/2011 : 16:04:24
|
John,yes in my opinion you can improve on your accuracy by setting up a base station. You don't need any extra kit.
A suggested method is to firstly check the machines to ensure they all perform in a similar manner because even machines from the same factory run will vary slightly and performance will thus vary. Checking them can be done as part of the process in setting up your base station.
Select the location of your base station. It should be situated to optimise ease of access and signal reception. To check your machines have them spaced far enough apart that they do not interfere with each other, 10 metre stations along a baseline should do it. You need to standardise the Garmins as best as possible so all should have same Garmin firmware, same make of fresh batteries, ideally from same factory run..do all you can think of to make machines as similar as possible including the orientation of antennae when recording. Make sure you have battery saver mode OFF, it smoothes data recording operation otherwise.
It doesn't matter next when you record data so much as how often you record and how many times you repeat. You can store the units in a clear plastic wrapper to protect against rain showers but avoid any day when heavier steady rain is forecast, no point in risking some snarl up. If you do protect the unit/s then be aware in theory the plastic will distort reception, very very slight degradation of data quality therefore. You then turn on the units and set them to record a continuous tracklog. I'm stuck here on how to select which of the three machines you have should be chosen for the base station and which for field, maybe others can advise but I'd opt for the two machines having most similar results even if the thrid might appear 'better' - what you want is standardisation even if that includes a systematic error. Select one of the machines as your base station unit and mark it so you never get it confused with the others.
Next select your permanent location for the base station again choosing best ease of access/ reception for the site. Run the base station Garmin recording a tracklog at the base station, download the data from it, remove any outliers in the data, average it. Repeat rinse and lather until you arrive at an average/median you know you’ll be happy with, the average/median then becomes your best estimate of the location of your base station, a sort of ‘fixed point’. I do all my stats stuff in Excel and I use GPS Utility for many GPS data functions even though there are free similar softwares out there. Do NOT use Mapmaker in any way shape or form as part of your data processing, you'll degrade the dat that way, you need to work on the raw data.
So for your field work a method would be to turn on your base station machine and leave it recording a tracklog whilst you’re out in the field. When you’ve finished, download the data from field and base station. You’ve already gotten a best estimate of the base station location so you can now look at the data for the base station for that day and see how it varies from the average (fixed point) at any time during the day. For any waypoint you’ve recorded in the field you can look at the time it was taken, compare that across to the exact same time of your base station tracklog for the session, see how far off from the ‘fixed point’ your reading at that time was and can adjust the position of the recorded waypoint accordingly. You can also try this with tracks using similar method. Or use a static track record as a waypoint. Either way will involve a lot of effort to even make a start on this let alone keep it operational and up to date.
As said these are all roughed out suggestions but hopefully they’ll be of some help. Whichever way you go if you do try some sort of base station out please do let us know how you get on, I'm sure many here would be interested in results. Good luck!
Mike |
|
|
|
djprang
36 Posts |
Posted - 26/08/2011 : 15:27:41
|
MM/Eric, there's a useful description of how one archaeology group uses Garmins in landscape survey here
http://www.swaag.org/publicationsSE2.htm
within which they go into the accuracy of the GPS units and how to optimise them for accuracy prior to recording and practically when recording waypoints. I notice they didn't seem to have the waypoint/data discrepancy between units you observed in your own trial so thought it'd possibly be of interest to you.
|
|
|
|
Philippe
France
295 Posts |
Posted - 02/09/2011 : 12:32:29
|
Hi John,
I've spent several hours (the least I could say) with my GARMIN 12XL testing Antonio Tabernero Galan 's ASYNC and GAR2RNX commands + GRINGO and P4 from the Nottingham University.
The precision of a GPS receiver is basically a function of the time spent at a given place...Sub-meter accuracy has price, would it be expressed in euros and/or hours!
Once googled, you could find it interesting to read : "Rapid Mapping with Post-Processed Garmin Data" by Chris Hill, Terry Moore and Michael Napier.
Links:
async and gar2rnx :
http://artico.lma.fi.upm.es/numerico/miembros/antonio/pd/
GPS Data Conversion (ie compress, crx2rnx, gpsdata, rnx2crx & help.txt)
ftp://epncb.oma.be/pub/software/rnxcmp/Windows/
Base stations data might be available here : http://www.epncb.oma.be/_trackingnetwork/stationlist.php.
Bon courage!
Philippe |
|
|
|
john parrott
United Kingdom
60 Posts |
Posted - 07/12/2014 : 11:00:06
|
I'm coming back to this topic because I now need to plot 1500 points to 1 or 2 metre accuracy, and I'm aware that technology has possibly moved on since my first enquiry.
Any thoughts on new hardware or software that might help?
John |
|
|
| |
Topic |
|
Improving GPS accuracy
Forum Locked
