GPS Precision Farming Accuracy

[dieselscout80]

I'll chime in on this thread as I work in the AG industry, as the company i work for builds many different sensors for Precision Planters.

Precision planting speed now is usually done by ground speed radar sensor. They also use a hall effect sensor looking at a wheel disk on the planter.

Planting also starts off in the office. There are several different PC software systems that tie all the information together. Each field is mapped using a GPS system.

The field 'prescription' is loaded into the tractors computer that communicates with the planter controller. Once the precision planter is in the field working it can then shut off or change rates to individual rows as it moves through the field. There is also a planter monitor that has seed population, target rates, %skips, %doubles for each row.

Some tractors are also equipped with 'Auto Steer' and auto turn on end row.

Dave

I have known about auto steering, but it is the first time hearing about auto turning at end of row.

Very cool!!!

 

[oosik]

The Environmental Health Dept ( Adams Co., WA ) had a Trimble setup when I worked there. They must have purchased it under some Federal grant program.

It was never used. By the time you unloaded all the gear - set it up - make the distance reading. You could make the same readings with a 100 foot tape - record the readings - be LONG gone.

I guess it was to record the location of septic systems - wells - buildings - property lines. Whatever - it was a total waste of time/money. At least for this purpose.

 

[spruce Deere]

I just asked this question on Reddit, but I probably should have come here first.

We're just a little berry farm using mostly mechanical tractors. Not a lot of electro hydraulic luxury, never mind computer controlled technology. What kind of accuracy do the big operators get with their gps precision farming setups. I mean my phone can't get within 10 feet of being right, but surely that's not close enough for row planning year after year. We'll never be big enough to need it, but I'm curious anyway.

Its not JD starlink, I use gps calc-an-acre set up, mostly for a speedometer when spraying. It works rather well for me.
The cal-can-acre has all sorts of pig tails for other implement adoptions and monitors such as seed drills and big spray rigs. I also can enter implement width in it and it will give me acres an hour, area selections and other things I may never use... but I have the technology if needed.

Dad used a radar deal that measured ground speed and figgered out stuff like the calc-an-acre. His is just a little quicker on 'start to read' speed, but appears just as accurate as my set up.

The more you spend, the more precision you get...

 

[Runner]

Cub Cadet has GPS "Autosteer" on their big commercial mowers. Sez I can mow .5 acre/hour faster with it....

SurePath Commercial Zero Turn Mowers | Cub Cadet US

Cub Cadet’s new SurePath auto steer technology utilizes GPS to produce the most efficient cut possible. Learn more here!

www.cubcadet.com

 

[SmallChange]

No experience with ag applications, but I got very interested in GPS as part of a hobby finding and surveying historical ruins in the woods. Mainstream individual GPS units, the kind many of us have in cars or handheld, do the best they can simultaneously calculating distance to each of the GPS satellites moment-to-moment and rolling this up into a single reported location. It's an amazing system, a great example of how excellent something very complicated can be. But there's a problem -- changing atmospheric conditions can make the speed of radio waves from satellite to ground vary a bit. Fancier GPS units measure and report "pseudoranges", meaning they don't combine all the distances into one result, they keep them separate for comparison with pseudoranges from other nearby GPS units. These can be reconciled later ("postprocessing", which I used to do), or even in real time. People often use CORS stations, which are high precision GPS units in permanent carefully surveyed locations that keep publishing their pseudoranges over the internet, for free to us users. Or, they use their own reference GPS units.

There are GPS units that deliver pseudoranges from the BadElf company for, IIRC, about $600 (this is several year old info). They'd be good for my surveying hobby using postprocessing. You have to have fancier equipment to do it in real time, which you'd need for steering a tractor.

Just fascinating, how they do all this.

 

[Texasmark]

It must be pretty good as the BTO farmers around here apparently use it as their fields are absolutely perfect with the plant location and color (fertilization) being absolutely even.

 

[LouNY]

It can be quite accurate but it requires a fixed location close by to get a correction signal from, with the correction signal they can be very accurate closer then an inch, without the correction station 8-12 inchs is all.
Then the higher end systems require two receivers one on the tractor and one on the implement being used.
As well as constant data links, cell phone or WFI. So a mere $12-16,000 can get you set with location service and then of course your equipment has to be upgraded to the latest and greatest to be able to do the varied seeding and fertilization. So heck for a million or so you can have your tractor and seeder set up real good till a sensor shuts you down for 2-3 days, interrupting a seeding or spraying or harvesting session which can easily end up being a 10% or greater production lose.

 

[mapper]

I believe top accuracy requires a calibrated (=location precisely determined) ground-based beacon in addition to the satellite cloud.

The system your thinking of is CORS, Constantly Operating Referance System, or something that has evolved out of it. This is a network of GPS receivers running on known control stations, their data is available via cellular connection. This allows for centimeter type accuracy on equipment in motion. When I started my career in survey and mapping this was the sort of technology we could only dream of ! I've been retired now for eight years, I'm sure that the technology and applications of it have continued to advance in the meantime.

 

[mapper]

There's a youtube channel called Cole the Cornstar, it's produced by a 26 year old farmer who's family farms 2000 acres in Iowa, they use this sort of technology, really amazing how precise these systems are. Planters and spayers with the individual planting heads and spray nozzles each separately controlled.

 

[LittleBill21]

so for those of you who use it, how is the reference station any more accurate, doesn't it suffer from the same ionospheric conditions?

 

[pmsmechanic]

so for those of you who use it, how is the reference station any more accurate, doesn't it suffer from the same ionospheric conditions?

From what I understand the inaccuracies come from the satellite signal being degraded for what ever reason. Probably military reasons but I don't know. The reference station signal isn't degraded and is only used by the owner and whoever he rents the signal out to. You can also pay for varying degrees of accuracy. You need subinch for rowcrop but 6" is good enough for general field work

 

[LittleBill21]

From what I understand the inaccuracies come from the satellite signal being degraded for what ever reason. Probably military reasons but I don't know. The reference station signal isn't degraded and is only used by the owner and whoever he rents the signal out to. You can also pay for varying degrees of accuracy. You need subinch for rowcrop but 6" is good enough for general field work

yea i am reading through it, reference station which is at a known point, sends the correction data to the local receiver, giving it the corrections needed to be accurate, but I have also seen units on a tripod away from the unit that is being used to measure, still trying to figure that out

 

[Williy]

YUP! You can build a tower and set up there and have a cool one while you watch your tractor roam around the field with you little control box

willy

 

[pmsmechanic]

yea i am reading through it, reference station which is at a known point, sends the correction data to the local receiver, giving it the corrections needed to be accurate, but I have also seen units on a tripod away from the unit that is being used to measure, still trying to figure that out

The little unit on a tripod is a portable reference station. Sometimes used for fields that aren't able to be reached by a stationary reference station.

 

[LittleBill21]

The little unit on a tripod is a portable reference station. Sometimes used for fields that aren't able to be reached by a stationary reference station.

so how is the portable reference station better? it doesn't know its exact location either, or is it generally a higher quality unit, that can provide better correction data?

 

[pmsmechanic]

I don't think it's better quality wise than a stationary reference tower. It's better because the tractor in the field can get signals from it vs the stationary tower that it can't get signals from. Most farmers around here have their own towers set up or pay a neighbor to use their tower. I don't see portable units much anymore.

Also the tower (portable or stationary) doesn't need to "know" where it is. The tractor triangulates it's position from a minimum of 2 satellites and a stationary tower and it knows where it is in relation to the tower and the satellites. That's all the tractor really needs to know to drive in a straight line.

Another fact to keep in mind is that the gps satellites etc. only send out signals. They do not receive signals from the earth. The tractor receives those signals and triangulates it's position from that information.

I don't 100% understand how it all works. What I do know from experience is that when it works it's great. When it doesn't work it's a pain in the rear. The tractor I drove all spring didn't have a working gps. So I relearned all the skills for driving straight that I developed in my 20's and got the spring work done. I call that skill eps. Ernie Peters steering.

 

[Torvy]

Cousins in the Black Hills use some of this tech. I watched them running the windrower that was equipped with GPS a couple of years ago. More accurate than the kid (25 ish) who had been working these same fields since he was 8 or 10. Once they set up, the operator is mostly just along for the ride. Monitors levels of drying agent and fuel and such.

It must be worth the expense. The ranch is run with scientific precision, but they also watch costs very tightly.

Anyone who hasn't seen a big family farm or ranch operation needs to try. It is truly amazing stuff.

 

[dodge man]

When I first started using GPS for land surveying we always had a second unit as a base station. This gives you accuracy of around one inch. We set the base up on a known point, which we set our own known points over the years but started from points set by the government over the years.

I never worried about different conditions between the base and the rover because it was fairly close, no more than a few miles away. A radio link connected the two.

Lately we tend to use an internet service to get corrections. We use a cell phone type link(a hot spot) to get the corrections. We pay for this service in Illinois and it’s through Trimble. Leica and Topcon also offer this. They have set up bases all over the state so you are using this network to work off. Obviously there is a lot of software and hardware development to make this happen.

As a rule different conditions between bases and rovers aren’t a problem, an exception might be you have heavy rain and storm over yourself and it’s small and isolated.

A little off topic but something that has improved gps is all the constellations. You have gps (United States), Galileo (European), Glonass(Russian) and Beidou(Chinese). When I first started using gps we had to do mission planning and maybe there was only about half the day there was enough satellites to work.

 

[SmallChange]

how is the reference station any more accurate, doesn't it suffer from the same ionospheric conditions?

The reference station does, but that doesn't matter. The fact that it is stationary and nearby means that the atmospheric errors will be almost exactly the same for both, and can be subtracted from the one you care about.

 

[dodge man]

One thing that I don’t know about ag users if they get indicators of the quality of the gps results. Our surveying equipment gave us indicators of the results and I always kept an eye on them. Even with the most modern equipment you still can’t get survey quality results in wooded areas.