Zero Turn Lawn Mower project

So I have about 15 acres probably more to mow and i have seen people using ardupilot to do so. I have quite a lot or arduino/raspberryPI experience and i am a machinist… but nothing with a drone. The mower is a zero turn type mower, i have already size up the linkages, got a good idea of how physically i will connect my large hobby servos, however as i have kinda tried to map out how i can do this… and i got some questions.

1 - can i power my ardupilot board with some 12v marine batteries wire in parrallel? I am trying to figure out the most effective way to do this because it will be mowing for quite some time?

2 - i have purchased 70 kg 12v hobby servos and they have 2 signal wires that go to the ardupilot gpio board, and two power wires (12v and ground) can i just power everything off of these 12v marine batterries?

i am going to just start with that and see if i can get a response. I found other threads talking about zero turn lawn mowers and ardupilot but no photos or wiring diagrams, just people talking

Thanks, Wallee

There are some very good reasons you don’t see any comprehensive guide to building such a monster:

  • Everyone’s project is slightly different. There is no cookie cutter recipe, and all of the differing brands/designs of mowers have different wiring/relay layouts, safety features, and even wildly differing hydrostatic drives such that the torque required to operate them varies greatly.
  • Most of us spend more time building and mowing than detailing and documenting.
  • Differing budgets move the goalposts by vast amounts.

Why not use the mower’s own charging system to supply 12-14 volts, such that you are only limited by fuel in the tank? That’s what most of us are doing. Even cheap voltage regulators/converters are usually tolerant enough to noisy charging system input to generate clean/useful output, and most 12V servos will suffer 13-14 volts without issue.

The servos you got are likely underpowered. Here are some alternatives that some of us have used with success:

A couple of us have been looking at these as well, but I’m unaware of anyone using them, so cannot vouch for their performance. I do have a couple of the DOCYKE branded ones on the bench right now, and they are VERY strong with metal internals.

You will also want to start looking into RTK GPS and moving baseline configurations for determining heading.

With a background in machining, I think you’ll find the fabrication required is straightforward and simple - just a few linkages to interconnect with servos. If you’re comfortable with electronics as you say, the wiring shouldn’t be all that bad, either. You’ll just need a basic understanding of servos and relays. The real challenges will be learning a bit about precision GPS along with the intricacies of how ArduPilot controls everything and then tuning the firmware controllers to suit.

Lastly, whatever you think your budget is, double it right now. I’ll look forward to you telling me how wrong I am about that :smiley:

Yuri, is it still the case that a magnetometer cannot reliability be used for yaw measurement in the mower application due to engine or metal of mower and gps yaw is the only way forward?
I noticed a vid on youtube (bitdog) says he has RM3100 working well. But i got the feeling from the time you spent getting gps yaw working that compass must be a dead loss

@ktrussell and I were talking the other day about improvements to ArduPilot along with our own better understanding of tuning that might make magnetometers worth exploring for budget projects, but I think you’d be hard pressed to get the same performance as a moving baseline configuration, and I wouldn’t put much faith in a cheap magenetometer when it comes to whirling blades and combustion engines, if for no other reason, operator and bystander safety.

I hesitate to use absolutes like “the only way forward,” but as of today, I can’t recommend using a magnetometer on a zero turn lawnmower with a combustion engine.

Also, the time I spent a couple of years ago is no longer a time investment that is required for GPS yaw to work reliably with ArduPilot. I was using it at a time when it wasn’t even an advertised feature in the Rover branch, and testing in the Copter branch had been somewhat minimal. It’s fairly plug-and-play with today’s firmware.

1 Like

Hey Wallee!

Welcome to the Mower World. I, along with several others here, are routinely mowing 5 to 30 acres with our machines and have gained a lot of opinions in the process.

Just a few to start:

  1. You need RTK in some flavor. Plan on it. Most of us have our own system but I am also using NTRIP as a trial. No issues to date.
  2. Plan on a second GPS on the mower for yaw info. Forget the compass for now.
  3. Power drain from the servos, guidance system and telemetry are very minimal. Your mowers regular battery and charging system are more than adequate.
  4. Bite the bullet and go with big servos. How big (strong) depends on your mower’s hydraulic control system. No issues going stronger than needed. Real issues with going too small. Our experience using cheap ones is really hit-n-miss.
  5. Budget: Wide range depending on choices, your mower and your situation.
  • RTK system - $400-$700 for your own system. If another RTK is close and on NTRIP, cost is $0.
  • Telemetry System - $100-$400 (Two Holybro Sik 915Mhz radios, antennas and coax.)
  • Mower Guidance System - $1500-$2500 (GPSs, Flight Control, Power supplies, R/C system, Relays, Box, etc)
  • Mower cost and mods - ??? (Bigger tank? Better tires? Warning lights? Etc)
  • PC to run Mission Planner
  • Misc strippers, crimpers, wire, magnifiers, soldering iron, etc - $100-$300
  • Unless IT is your background, don’t under-estimate the amount of reading/learning involved. No worries though. It’s not going into space and anyone with patience and desire can certainly build one. But getting a real mower sorted and working properly is not a weekend project.

The feeling of accomplishment when the animal actually works? Pretty cool!

Bring your questions. Onward!

-Steve-

1 Like

Thanks yuri, i spent yesterday learning about RTK gps, very interesting i can see why that is a must have. You said in your response that i could use the 12v coming from the mowers charging system, how exactly would i do something like that? i understand the alternator is putting 12v back into the battery, but would i wire directly off the terminals of the battery? or do i need a voltage regulator? i am pretty good about doing my research, but it never occurred to me i could even do that.

And thanks for all the responses

It’d help if you could get a wiring diagram for the mower you intend to retrofit, since you’ll probably want to tap into multiple circuits for safeties/features.

The easiest way to get 12V is straight from the battery terminal, but be sure to use a fuse and probably at least one switch as well if you tap the power so directly. You could also locate the fuse block and use some existing circuitry there, so long as you don’t overload any given circuit.

Maybe pick up a book about basic automotive wiring. Your questions indicate a pretty shallow depth of understanding. Mowers and cars share a lot of fundamentals.

Hi Yuri_Rage

Thanks for the information that you have provided. Just looking at the servos and wanted to ask if the response time 0.5 seconds to 60 degrees is acceptable. I notice the two that you have used or know that have been used successfully are very responsive. 0.08 seconds to 60 degrees.

Thanks in advance for your feedback.

Matt

I wouldn’t use a servo that slow. You probably don’t need something 5 times faster, but 2-3 times faster is likely warranted.