I’ve been farming with my father in Jamestown, South Australia for years. Mixed operation, cropping and livestock. Like every broadacre farmer, I know the value of straight lines. And like every farmer who has looked at the price tag on a proper guidance system, I have had that moment where you think: there has to be a cheaper way to do this.
The problem
The awkward thing about GPS guidance is that the basic building blocks are not the expensive part. Receivers, antennas, laptops, serial data, and A-B line math are understandable and available. But a basic commercial guidance setup can still push you into proprietary displays, dealer installs, subscriptions, and hardware ecosystems that are hard to repair or adapt.
That ecosystem has value for some jobs. But not everyone needs all of it all of the time. Sometimes you want a practical tool that helps you drive straighter and lets you keep your own data.
The spark
I’d been teaching myself to code for a while: Python first, then Rust when I wanted something faster and more reliable. I was building other tools for the farm, and one day it clicked: I already know what a lightbar needs to do. Why not build one?
The first version was deliberately simple. Read GPS over USB. Parse NMEA. Show the position. Set two points. Calculate cross-track error. Draw a lightbar.
Within a month, I had a working guidance tool.
What Finn Guidance does
The public idea is still simple. You have a Windows laptop. You plug in a USB GPS receiver and antenna. You run the software. You set an A-B line by marking two points in the paddock. The lightbar shows how far off-line you are in real time.
The details matter:
- Auto-detection - plug in the GPS and the software tries to find it.
- Field management - save A-B lines grouped by field and load them again later.
- Auto-pass selection - set your implement width and the software selects the nearest pass line.
- Nudge - offset the guidance line by a precise amount for inter-row or alignment work.
- Coverage logging - record where you have been and export the data.
The field project has also grown past the first public-lightbar story. The tractor direction now uses an LC29H BA receiver direct to the laptop, with a motor ESP32 handling the steering inner loop. That is field-prototype work, not a finished public product.
What it costs
The software is free. Open source, no strings.
The hardware is not the old “thirty dollar receiver” story once you include the antenna and a setup you would actually mount on a tractor. A realistic budget for the GPS module plus antenna is around $100 AUD. Cables, mounts, weather protection, power, and steering hardware are separate.
That is still far below the cost of most proprietary systems, but it is better to be honest than catchy.
What it does not do yet
Finn Guidance’s public release is a guidance/lightbar tool. It tells you where to go; you still steer. With standalone GNSS and no RTK correction service, expect metre-level behaviour. That can be useful for broadacre work, but it is not controlled traffic accuracy and it is not a certified auto-steer product.
RTK support is part of the FINN Base direction. FINN Pilot is the next implement-awareness piece: outside-tine GPS, shaft speed, safety switches, camera capture, and eventually coverage logging that reflects what the implement is actually doing.
Why open source?
Because I have been on the receiving end of proprietary ag-tech that stops being supported, gets acquired, or requires a subscription that keeps going up. Open source means the code is there for anyone to inspect, modify, or build on.
It also means other farmer-developers, students, and tinkerers can contribute. The best features will probably come from people solving problems I have not thought of yet.
What’s next
FINN is becoming more than one app:
- Finn Guidance for the tractor and cab.
- FINN Pilot for implement sensing and future coverage authority.
- FINN Base for local RTK corrections.
- FINN Core for field-run records and analysis.
- FINN Interface for operator interaction.
- FINN Copter for mapping and future farm-state updates.
The goal is not to merge all of that into one fragile program. The goal is a set of practical farm tools that can stand alone in the field and connect when there is value in sharing data.
If you want to try the public guidance path, start with the Finn Guidance page. If you want the hardware reality first, read the Hardware notes.