[youtube=http://www.youtube.com/watch?v=LDzHwy28kng]
We’ve seen the Arduino board in charge of some pretty unique tasks in the past. Harvesting locally grown soybeans was not one of them.
[Lance] rigged this beast up in order to automate the monotonous task of driving up and down the vast soybean fields of Iowa. The 15 ton farm combine’s hydraulic steering pump is at the mercy of a team of gadgets, including a GPS, Windows 7 PC, and the omnipresent Duemilanove (which acts as the output card, connecting the PC to the pump). So far, it is reported to be doing a great job, straying only about an inch and a half from its desired, GPS-programmed, path. Even if the Arduino decides to go totally berserk and drive the combine off course, speeding around at 5mph makes it pretty avoidable. A supervisor is also in the cabin at all times, looking out for errors. [Lance] eventually hopes to offload all steering-related calculations to the ATmega328P onboard.
Commenters are welcome to share heavier-duty uses for the Arduino (if they exist).
This is pretty cool, the systems already exist, but they cost £££££££/$$$$$$$.
Someone at CERN should make the LHC tweet through an Arduino(wait, it doesn’t do that already!?)
Hmm i guess this guy has skipped a few steps:
“I’m an arduino noob, and this is my first project using an arduino”-[Lance]
I’ve seen systems like this too. They cost A LOT.
A lot of things are like that…stuff that is actually pretty simple to build, but could have horrific consequences if something went wrong. Therefore companies have to spend a lot of money on testing, certification, redundancy, lawyers, etc. That drives up the price a lot, though if you’re running a business it can be worth it not having to do all the testing yourself.
Kind of like new medicine, the formula may be simple enough for someone to make in their kitchen, but it requires years of FDA testing before it can be out there saving lives.
Nice project, but I’m left with why? There is no indication the investment lead to more income at year’s end. While this is probably less expensive( depending how the final cost is calculated) than commercial units, why spend the money, when there’s an operator in the cab. sorry for the upcoming rant, but it’s December. The month I pay higher property taxes, and pickup registration, to subsidize the lower property taxes and registration of farmers/ranchers in KS.
Yea I know part of the why is that he could do it. I would like to see a more detailed account of the project than reading through a forum. This doesn’t use the everyday consumer GPS receiver. Does this individual have to set up his own RTK base station or do farm States like Iowa have a free to use network of RTK transmitters set up? Any details Lance may have reveal may be spread across multiple forums, and God how I hate wading through forums. :(
The biggest point of the above post is that youre already paying someone to sit and monitor for errors. This means that the guy essentially has to be skilled enough to drive it, too. So then why the arduino setup? I suppose it makes his job easier…
To answer the why question: It does save time. When you have a 60 foot boom, then the operator needs to overlap at least a foot or more to make sure he doesn’t miss any crop. With a gps unit, you can drop that overlap to about an inch or two. Might not sound like a lot, but when you have 5 miles of crop to harvest one pass at a time, that smaller overlap can save you a lot of trips.
In NW Iowa, there are RTK base stations in just about every town. Given that towns are normally 10-20 miles apart, I am close enough to be able to choose between 3-4 RTK base stations per field. All I do is input a code, which I’ve never had to pay for because the base stations are either owned by the local John Deere stores who many do business with, or friends that put up base stations themselves.
And, there are enormous benefits to using precision farming techniques like this, although I agree auto-steer doesn’t do much other than free up hands. One example is that you can plot and save yield maps and compare these yield maps to the variety of crop planted in the spring and the type of fertilizer used, etc.
Aren’t all crops “locally grown”?
Auto steer helps to reduce driver fatigue. Less fatigue means longer hours. Longer hours means fewer days, and some years, like this one, nature does not give you many days to harvest your crops. (I work for a CaseIH dealership in West Memphis, Arkansas)
why=productivity
with auto steer you can run really wide harvesters. and use the full width of the header. like the CAT Lexion 590r
Yup, with manufacturers pushing up to and beyond 40 foot headers now on harvesters, this is becoming more and more useful for improving productivity by reducing operator error.
Up until now, GPS guided steering was largely only really needed/used for planting and spraying applications where rows need to be equally spaced (planting) or the widths were incredibly wide (spraying).
We use lots of precision Ag stuff on our farm. The primary reason we have it is to have long term maps of our crops. Not only are you able to record yield, but you can also grab as applied maps for seeding rate, fertilizer and other products you are putting on. This allows you to easily conduct informal studies and see the affects. The price point on comercial systems has fallen such that it is almost the same price to put physical markers on a planter as to purchase a GPS auto guide system.
Maybe generation 2 can have enough safeguards (perhaps something like the remote E-stop system used in the DARPA autonomous vehicle competitions) so that one operator can monitor multiple units.
If it’s not already, this would be a great open source effort.
Not to mention the driver only has to supervise. He can read, or do something else while attending… that he otherwise wouldn’t be able to do.
Kalman filtering?! Why the heck are you farming dude? Go get an engineering job :)