Predictive maps help weed control

Greg Stewart is different from most innovators in Canadian agriculture.

People that start ag companies are often plant scientists, agronomists or experts in farm machinery. Stewart has a master’s degree in applied mathematics from Dalhousie University and a PhD in control engineering from the University of British Columbia.

“A controls engineer ensures that an organization can create high-quality products in the most efficient manner possible,” says coursera.org. “There are many different types of control systems, but each one serves the same purpose: to control outputs.”

Stewart, who lives in Vancouver, has used his expertise in a range of industries, including pulp and paper, automotive powertrains, semiconductors and designing brakes for airplanes.

He is the founder of Geco Engineering, which has turned its attention to crop production.

“My passion is in developing and deploying advanced machine learning (or AI), control, and automation technologies … to address meaningful problems,” says his UBC profile page.

“I’m fascinated by the challenges and opportunities posed by agricultural problems.”

Right now, he’s focused on the problem of weed management.

Most farmers apply a single herbicide or multiple herbicides across an entire field to control weeds. Stewart has developed a technology that creates predictive weed maps so farmers can use herbicides more strategically.

“We’re making predictions about where weeds will emerge. The technology is a combination of AI (Artificial Intelligence) and agronomic modelling,” Stewart said. “We’re looking at what the weed population has been doing over the last five years… Then we’re able to make a prediction about where they will appear next year.”