Blue Proton Initiative: four 17-year-olds are building AI-powered livestock monitoring with Arduino

Pietro Maria Piazza, Alessandro Nesci, Davide Santucci, and Matteo Angiolillo are not waiting to finish school before starting to build something real. Based in Forlì, Italy, the four friends behind Blue Proton Initiative strive to develop an AI-powered livestock monitoring system designed to help farmers identify individual animals and detect early signs of health problems – automatically, in the field, in real time.

They are 17. The project is four months old. They are far from ready to start their own business, but they sure have the spirit of budding innovators – and they are already pushing the Arduino® UNO™ Q board to its limits.

Agriculture 4.0 can’t wait

Livestock monitoring an area where technology has lagged behind what’s actually possible. Keeping track of individual animals, spotting health issues early, flagging anomalies before they become costly problems – most of it is still done by hand, which means it’s slow, inconsistent, and easy to miss.

Once they identified this gap, “None of us wanted to wait,” the team explains. “The tools, the knowledge, and the resources needed to tackle real problems are accessible today, and we saw no reason to postpone.”

What they want to build is a system that uses computer vision and custom-trained neural networks to identify animals and monitor their behavior – the kind of continuous, low-overhead monitoring that would be genuinely useful on a working farm. “In farming, catching something late can mean real losses. An automated system that runs in the background without needing constant attention is genuinely useful in that context.”

Four teens, one dream team

The four friends have divided responsibilities: Pietro leads software engineering, Matteo develops the backend using C, Davide handles CAD design and the physical structure of the system, and Alessandro focuses on neural network development and YOLO model training – the core of the system’s ability to detect and count animals automatically.

The models are being developed entirely in-house. Dataset composition is central to their approach: the more varied the training data – different lighting conditions, animal movement, weather, environments – the more robust the model becomes when deployed in real-world settings. As Alessandro puts it, “with enough diverse images covering a wide range of real-world scenarios, the model learns to generalize.”

UNO Q sits at the heart of the architecture. The entire system is housed in a custom 3D-printed enclosure designed specifically around specific requirements, keeping things compact and field-ready. The team is writing and deploying code directly via the Linux terminal – no IDE hand-holding, no App Lab. Just the board, a text editor, and a lot of iteration.

Why choose UNO Q?

Blue Proton didn’t come to Arduino cold. The team had already used the UNO and Arduino^® Nano^™ hardware in earlier projects – a robotic arm, a model rocket – but they were quick to recognize that UNO Q was something different.

“The ability to run Linux on a compact, affordable board was a turning point for us. That combination – processing power, lightweight form factor, and low cost – made it a natural fit for a project like ours, where the hardware needs to operate in the field, potentially in remote environments, without being bulky or expensive to deploy.”

The ability to run AI inference locally, without depending on a constant server connection, is essential for agricultural deployment. UNO Q makes edge AI viable within real student-budget and field-deployment constraints.

That said, the team is candid about the hardware’s tech specs: “RAM and CPU are the main bottlenecks – running a system like ours on a board with those constraints requires every component to be as lean as possible.” Their approach: test, optimize, test again. “There is no shortcut to it – it just takes time and a lot of patience.” 

(Of course they started on the project before the launch of the Arduino® VENTUNO™ Q board, which would allow the Blue Proton Initiative team to have more headroom!)

Innovation is all about the journey

Four months in, Blue Proton has a working cattle-counting system, a clear technical direction to develop the health monitoring component of the project, and a level of honesty about their process that’s refreshing. “We are still working on it. We are 17, we have a math test tomorrow, and we are working on something that could maybe become our job one day – or maybe not. We don’t have a commercial product ready, and we are not going to pretend we have a five-year plan. But we are going to keep building, keep learning, and see where it goes. That is enough for now.” 

Their advice to other students thinking about starting something similar: “If you have an idea, chase it. Do not wait until you feel ready, because that moment will never come. You figure it out as you go, and that is kind of the point.”

Follow Blue Proton Initiative’s progress on their website and Instagram account – they are producing video documentation of the build as it develops. 

Arduino, UNO, VENTUNO, and Nano are trademarks or registered trademarks of Arduino S.r.l.