Recommend GPS Unit

[imoffit]

I was curious if anyone has any recommendations on GPS units that they use to mark GCPs and verification points. I do not have the money for a survey grade base and rover type GPS unit but I am hoping for something accurate enough for jobs involving volume measurements and agricultural mapping.

Thanks!

[Saijin_Naib]

What is your budget, and what is your accuracy need? These are going to put some very hard constraints on what are options you can realistically use.

[Ivan_Gayton]

If you are up for some stuff that’s more technically challenging than a polished commercial GPS unit, ArduSimple (https://www.ardusimple.com/) sells a board with a uBlox ZED-F9P dual-frequency GNSS receiver, as well as antennas and radios. You can get a base and rover for less than $700 all inclusive, which as a set will easily get you 5cm horizontal and 10cm vertical accuracy (better if conditions are good and you know what you are doing).

We’ve had very good results with this in Tanzania. It takes a fair bit of fiddling and patience to get it going though. This GitHub - hcwinsemius/RTK_GNSS: Manual for RTK GNSS with low cost u-blox ZED-F9P receivers rover-base setup might give you a sense of how complicated it is, and whether it’s something you want to take on (though all of that stuff was written quite a while ago, and the process has become a bit simpler since then).

[daniel.dobosi]

This is really interesting stuff, thanks for sharing!
An emlid Reach RS+ is in the price range of 800 USD, could you please explain the difference and benefits over ardusimple? I did a quick internet search and they cost both the same approximately, even if you build yours?
Thanks!

[Ivan_Gayton]

The Emlid Reach RS+ is about $800 for each unit, and if you don’t have access to a base station you need two for high accuracy.

The ArduSimple is about $300 for each unit including antenna (call it $400 if you go for higher-quality antennas). Again, unless you have access to another base station for corrections, you need two. So about half the price.

To oversimplify a bit:

• Both will give similar (excellent) results if used properly
• The ArduSimple is way cheaper
• The Emlid takes way less fiddling

So it probably boils down to whether you are more constrained by time & expertise or money.

In my case, I also happen to like the mostly-FOSS approach of ArduSimple, but that’s a philosophical rather than practical concern.

[Saijin_Naib]

If you can get away with lower accuracy thresholds (and maybe run PPK after collection), I’ve had great luck with the BadElf units like these:

Bad Elf GNSS Surveyor $650:

Bad Elf GNSS Surveyor

Bad Elf GNSS Pro+ $300:

Bad Elf GPS Pro+

Depending upon your phone, you could always do some long-baseline/point averaging with something like GPS Point and return sub-3m accuracy pretty easily:

Using GCPs in XYZ format

Excellent as always, Isreal. If you have Android and a bit of patience, GPS Point/Position Averaging (long-baseline sampling) goes a LONG way towards improving over standard single-sample accuracy. I like this app: GPS Point - Apps on Google Play Sitting on my desk indoors in my apartment, no clear view of the sky from an entry/mid-tier MediaTek SOC Phone: [image] That is to say, you can do much better with applying a few simple GPS collection methodology things you’d normally do.

[daniel.dobosi]

Great stuff!
So the if I understand correctly, the op is talking about PPK. That means no rover but only one base station for measuring GCP’s. So we have mentioned

• Bad Elf with sub 2.5 meters accuracy for 650 USD
• Reach rs+ without pole for 800 USD
• and finally the ArduSimple for 444 USD without pole + free android app “SW maps” for the GCP logs
  right? From the price/performance ratio that’s a bonus for the ArduSimple.

Thanks!

(ofc not to mention the DJI stuff and the Reach rs2)

[imoffit]

At the moment those units are likely to be a little to expensive for me. What is the minimal degree of accuracy that is required for things like mapping crops fields or performing volume estimations of things like gravel piles at construction sites?

[Saijin_Naib]

@imoffit

What is the minimal degree of accuracy that is required for things like mapping crops fields or performing volume estimations of things like gravel piles at construction sites?

Unfortunately, that is something you need to decide. If you have had talks with your customers, you can decide what they need/want, and set your product standards and pricing accordingly.

From my work in precision ag, I’ll volunteer this: Almost nobody has per-nozzle/per-row variable rate equipment (including things like GreenSeeker). What that means is they’re doing variable rate per implement, which could be something like 20-40m wide depending upon the size of the farm/field. The difference in a meter or two with a setup like that is completely negligible.

However, stockpiles can be more tricky. Again, what is their expectation, and can you reasonably match it with the standards you set for your own products?

[smathermather]

@imoffit

At the moment those units are likely to be a little to expensive for me.

The below might be your best option then. Once you get into the 30-45 minute range of averaging, your accuracy is quite good:

@Saijin_Naib

you could always do some long-baseline/point averaging with something like GPS Point and return sub-3m accuracy pretty easily:

Excellent as always, Isreal. If you have Android and a bit of patience, GPS Point/Position Averaging (long-baseline sampling) goes a LONG way towards improving over standard single-sample accuracy. I like this app: GPS Point - Apps on Google Play Sitting on my desk indoors in my apartment, no clear view of the sky from an entry/mid-tier MediaTek SOC Phone: [image] That is to say, you can do much better with applying a few simple GPS collection methodology things you’d normally do.

[uplanddog]

Ardisimple sells an option that can used with NTRIP services, might be an option if available in your region.

[geoma]

Is there an advantage of going with these devices instead of a Garmin GPSMap 65s?
I have one of those Garmin, they have a feature for averaging waypoint measuring, wo you basically stay a good while on one point and it increases precision by averaging multiple measures taken at the same spot.
https://www.garmin.com/en-US/p/715072#overview

I have never worked with GCP before, would it make sense to use GCP with this Garmin GPS?

thanks!

[Saijin_Naib]

I’m not sure the Garmnin GPSMAPs output raw RINEX measurements so you can do PPK like you can with the Bad Elf units. I’m also not finding their rating for expected accuracy like they used to rate their older products with.

I have not had the chance to do a head-to-head comparison of GPS averaging on the newer Garmin units vs the Bad Elf vs a regular cellphone, so I can’t comment here.

GCPs are almost always a good idea for the reconstruction, even if you need to use your phone (and do some “short” point averaging [60s is a big jump in quality from a single reading]).

[AgingUETiger]

Check my evaluation of the Emlid Reach M2 Multi Band GNSS Receiver here. You can use as a standalone receiver if in proper range of CORS base stations for post processing or in cell phone range for real time NTRP corrections in kinematic mode.

Reach M2 w/ Mini-Survey Antenna Static/Rapid Static Testing Results