A group of Australian farmers is developing an artificial intelligence tool that learns how an individual farm works and takes over much of the record-keeping that eats into farmers' evenings, Beef Central reported.
The tool, called Farm Assistant, was built by Regen Farmers Mutual, a farmer cooperative formed to deal with environmental markets. It says just 3% of Australian agriculture, forestry and fishing businesses use AI, the second-lowest rate of any industry and well below the 12% national average.
Co-director Andrew Ward said low uptake should not be mistaken for reluctance, since farmers have used GPS, automated machinery, sensors and remote monitoring for years. "They don't need AI to tell them how to farm," he said. "What they do need is a way to cut down the hours spent at night entering information into different systems and doing paperwork."
Farmers record what they are doing as they work, by voice note, photograph or text from the paddock, tractor or yards. A farmer treating livestock can tell the assistant what was done, and treatment records and inventory are updated and paperwork prepared without entering the information again. The software learns the farm's own language, from paddock names to livestock groups. "If you've called a paddock 'The Back Forty' for 30 years, you shouldn't have to rename it because you've bought some new software," Ward said. It is designed to work alongside farm-management software, weather data, machinery systems, sensors and commodity data.
Twenty farmers are piloting the tool and meeting weekly in what the cooperative calls an "AI gym", and more than 40 farms are waiting to join. Co-director and cattle farmer Christophe Bur said the aim is to stop information disappearing into notebooks or memory: "If AI means someone gets through the gate at the end of the day and has dinner with their family instead of sitting at the computer for another two hours, that's technology solving a real problem."
Photo: Cgoodwin / Wikimedia Commons (CC BY 3.0)
Source: Beef Central
Comments
(0)