Artificial intelligence could help future farmers after cucumber competition lands Harrow scientist a win

A team of researchers that won an international artificial intelligence (AI) and agriculture competition included a Harrow, Ont. scientist with Agriculture and Agri-Food Canada.

Team Sonoma grew more than 55 kilograms per square metre of cucumbers to secure victory 

A team of researchers that won an international artificial intelligence (AI) and agriculture competition included a Harrow, Ont. scientist with Agriculture and Agri-Food Canada. (Wageningen University & Research Centre)

A team of researchers successfully grew a crop of cucumbers in the Netherlands — without being there.

Some of that team included a scientist based in Harrow, Ont.

Scientists from Microsoft Research AI, the Ontario Ministry of Food, Rural Affairs and Agriculture, as well as Danish and Dutch university students monitored the crops the same way you would monitor a crop in your backyard — except they used artificial intelligence from remote locations. 

It was part of the first annual Autonomous Greenhouse Challenge, which the Microsoft team won. 

Xiuming Hao is a greenhouse crop physiologist and energy conservation expert with the Harrow Research and Development Centre. 

Hao's group, Team Sonoma, succeeded in winning the challenge by growing over 55 kilograms per square metre of cucumbers — controlling elements in the greenhouse from right here in Harrow. 

Cucumbers were grown and monitored using artificial intelligence. (Wageningen University & Research Centre)

"We basically controlled all the greenhouse climate," said Hao. "The temperature, humidity, lighting, carbon dioxide and the irrigation. We remotely … controlled all the settings and [tried] to optimize all the inputs."

The competition was organized by Wageningen University in the Netherlands; was sponsored by Chinese tech giant Tencent; and ran from August 2018 until the first week of December. 

Five international teams were tasked with growing the most cucumbers in a 96-square-metre compartment while minimizing the amount of water, carbon dioxide and energy used per kilogram — in the most technologically innovative way possible.

According to Agri-Food Canada, Team Sonoma's net profit — the sale price of the harvested cucumbers minus the cost of energy, water, labour and other resources — was 25 per cent greater than the second place contenders. Compared to other cucumber growers in Essex County, the harvest was considered "high yield."

Hao said that it will likely take two to three years to implement his team's research in a non-lab setting and that the system is currently geared toward helping growers make more efficient decisions.

"This is not going to … replace the manual worker or the general worker yet," said Hao. "That part will be coming later when we get into … greenhouse automation with robots."

Hao said the system will also have the capacity to get better and learn year-after-year, which would include expanding into other crops. The cucumber came first because it had the most data to get started. 

Though there was no monetary prize associated with the competition, Hao said the real reward is being able to develop new algorithms capable of improving future crop production.

"The prize to me is to have had this opportunity to share my knowledge and combine with the latest technology," said Hao. "This can help the future farmer."

The next autonomous greenhouse challenge will require participants to grow and maintain a crop of cherry tomatoes.

With files from Windsor Morning


To encourage thoughtful and respectful conversations, first and last names will appear with each submission to CBC/Radio-Canada's online communities (except in children and youth-oriented communities). Pseudonyms will no longer be permitted.

By submitting a comment, you accept that CBC has the right to reproduce and publish that comment in whole or in part, in any manner CBC chooses. Please note that CBC does not endorse the opinions expressed in comments. Comments on this story are moderated according to our Submission Guidelines. Comments are welcome while open. We reserve the right to close comments at any time.