U of G Engineering Lab Set to Revolutionize Greenhouse Vegetable Harvesting

Thu, 27, June, 2024 by Food from Thought

At sprawling greenhouses in the rural southern Ontario community of Leamington, one-armed robots designed and developed at the University of Guelph roam rows of tomatoes and gingerly pluck them from vines with a mechanical hand.

It’s a process that takes about 15 seconds, but a team led by Dr. Medhat Moussa, a professor in the School of Engineering at U of G, hopes to cut that nearly in half, making it comparable with the work of skilled human harvesters.

This would be a game-changer for Canada’s greenhouse industry, potentially saving growers millions of dollars a year and helping mitigate a looming labour shortage.

“This is a pain point that has been around for some time,” said Moussa. “The cost of labor is quite high, and at the same time, these are perishable items. You need to harvest and ship, in some cases, massive quantities, so any disruption of labour will have an impact on your production.”

These tomato-picking robots, part of the Guelph Intelligent Greenhouse Automation System (GIGAS) developed by Dr. Moussa’s robotics lab at U of G, are set to begin sustained long-term field testing this summer with three partners in Leamington and Guelph. This research was undertaken thanks in part to funding from the Canada First Research Excellence Fund, with support from the Ontario Agri-Food Innovation Alliance, a collaboration between the Government of Ontario and the University of Guelph.

This is a crucial step toward commercialization, potentially leading to the birth of a multi-billion-dollar industry. Each robot is valued at roughly CDN$100,000 and greenhouses would likely require one robot per acre, said Moussa. With about 4,000 acres of vegetable-growing greenhouses in Canada (roughly two-thirds in Ontario, one-third in British Columbia), the potential market is huge.

“We expect that two years from now the technology will be ready for commercialization,” said Moussa. “The business model might take another two years, but I’m trying to bridge the gap to commercialization right now.”

If successful, the GIGAS may be part of a solution to the anticipated retirement of a huge portion of Canada’s agricultural workforce by 2030, leaving more than 100,000 job vacancies. Greenhouses are a key industry segment and would feel the shortage acutely.

The labour savings from automated harvesting would be considerable, since robots (despite currently being slightly slower at picking than humans) can work longer hours than human labourers.

An even greater benefit for the industry would be avoiding labour disruptions in pandemics like COVID-19 when greenhouse production fell by about 30 percent in Canada, said Moussa. That translated to about $150,000 per acre in lost revenue, in part due to quarantine requirements for human workers.

GIGAS envisions a world in which robots complement human labourers and strengthens the value chain by completing many other tasks beyond simple harvesting.

The same robots could also use digital cameras to scout for diseases and identify plants at the perfect time for harvesting, predict the yield and generate data that can guide shipping schedules.

They would also use machine learning to continuously improve harvesting processes.

“The value proposition is not just reducing the cost of labour, but improving your operation as well,” said Moussa.

Field testing will involve several tomato varieties, all of which have slightly different harvesting requirements; the plan is to then apply the same technology to bell peppers and cucumbers, potentially expanding later to apples and other fruit crops.

While helping address the agricultural labour shortage, Moussa said GIGAS could lead to significant job growth in the manufacturing and system integration industry. By improving food system efficiency and boosting grower profits, he also predicts Canadian greenhouse operations would expand and create additional jobs.

“The technology is quite complex, so there’s a number of components,” said Moussa. “We have now reached a point where you have a good prototype that can do the harvesting within a reasonable cycle time, and we can see a world in which that system can be built with a reasonable cost.”

“There are multiple spin-offs in the value proposition,” he added. “Changing the value chain that exists now, between producers and the end user, is not easy. This technology can enable it.”