• Sun. Nov 29th, 2020

Dimancherouge

Technology

Alphabet Mineral reveals crop-inspecting robots

A photograph of a field at sunset shows three bridge-shaped robots - a central bar suspended by two pillars - drive over rows of crops
The plant buggy roams atop the crops, counting and analysing each

Google’s parent company, Alphabet, has unveiled prototype robots that can inspect individual plants in a field, to help farmers improve crop yields.

The robot buggies roll through fields on upright pillars, so they can coast over plants without disturbing them.

The goal is to collect huge amounts of data about how crops grow.

Called Project Mineral, it is part of Alphabet’s X company, which aims to create world-changing technology from radical “moonshot” ideas.

In a blog post, project lead Elliott Grant said: “We hope that better tools will enable the agriculture industry to transform how food is grown”.

A modern version of the buggy rolls through a huge field of low-height green crops, stretching as far as the eye can see
The robot buggy can be made in different sizes for different crops and planting patterns

The team says its main goal is to address the world’s increasing need for food and the sustainability of growing it.

But current tools do not give farmers the kind of information they need.

“What if every single plant could be monitored and given exactly the nutrition it needed?” Mr Grant wrote.

“What if we could untangle the genetic and environmental drivers of crop yield?”

A man works on a prototype buggy in a field at night - this version looking like a large table mounted on bicycle wheels, with computer kit on the flat surface
The latest version of the buggy is much more polished than its prototypes

While farmers may have information about the soil content or the weather, the buggy robot was designed to see how plants were “actually growing and responding to their environment”, the company said.

“Over the past few years, the plant buggy has trundled through strawberry fields in California and soybean fields in Illinois, gathering high quality images of each plant and counting and classifying every berry and every bean,” it said.

On top of being a literal bean-counter, the buggy can also record information such as plant height, leaf area and fruit size.

And all that data is plugged into a machine-learning system to try to spot patterns and insights useful to farmers.

“Growers can troubleshoot and treat individual plants instead of entire fields, reducing both their costs and environmental impact,” the company said.

Mineral said it was already working with breeders and farmers in Argentina, Canada, South Africa, and the United States.

But there is no explicit timeline or plan to release the buggy as a commercial product.

Source Article