Advances in fruit and vegetable growing and packing are set to evolve further in the near future. As ANDY WILSON reports, data processing is anticipated to be first task of artificial intelligence.
La Trobe University hosted food producers, innovators and the generally curious to its ‘Navigating the Future’ talk at the Shepparton Art Museum in October.
A panel of three experts all agreed that the most beneficial use for artificial intelligence in food and vegetable production would be with analysing the enormous amount of data collated by growers using automated systems.
La Trobe University crop and plant science lecturer Associate Professor Tony Gendall said the biggest teaching challenge for him was encasing all applications of technology within a short education time frame.
SPC engineering manager Mark Hunter spoke about the use of technology in their cannery, including the monitoring of fruit colour for quality and the use of a mobile robotic dog to measure conveyor vibration and temperature.
Flavourite operations manager John Murphy manages 75 hectares of vegetable farms across Victoria, concentrated mostly in the Goulburn Valley.
John said initial use of AI in glasshouses in Holland had proven unsuccessful.
“So, our first step is pest and disease management, getting the growing taken care of and collecting the data on plant health,” he said.
“We spend a lot of labour on monitoring our crops, so to be able to actually integrate that data into production forecasting will have a massive impact on waste and sales.
“Data interpretation models are … probably what we really need.”
Tony said the COVID-19 pandemic allowed for advancing general DNA sequencing technology to an “unimaginable” scale.
“Data is something that is emerging and it’s going to continue to grow and it’s something that AI is actually meant to help us with it,” he said.
Mark said some technology companies were putting the AI ‘banner’ on what he considered non-AI developments.
“The new companies (are) calling us saying ‘hey, come and try this’,” he said.
“And a lot of the time what we see is probably not what I call AI, it’s just automation.
“I also think the data processing side is where we’ll see it come through first.
“I’m not sure we want the AI to make a decision that we definitely act on but make a suggestion only.”