May 9 2025 - Shaun Sharpe, AAFC
Quick recap
Leia introduced herself as a new member from Nophia, while Shaun Sharp from Agriculture and Agri-food Canada presented on physical weed management principles and applications for precision agriculture. The discussion also covered the challenges of species identification in weed detection and management, the potential of cloud and edge computing for image analysis, and the need for more studies on the economic viability of different weed management solutions. The conversation ended with announcements about changes to the agendas, the Agrotech Venture Forum, the Southwest Food Innovation Summit, and the invitation for companies to showcase their technology at the demo day.
Next steps
Summary
Ag Tech and AI Advancement
In the meeting, Leia introduced herself as a new member from Nophia, sharing her experience in implementing ag tech GIS software and her interest in advancing ag tech and AI. Vivek Rajakumar from Picket was also present but did not have the chance to introduce himself. Michael, an assistant professor in computer science at the University of Winnipeg, introduced his research group's work on robotics for data collection and machine learning applications. The main presentation was given by Shaun Sharp from Agriculture and Agri-food Canada, who discussed physical weed management principles and applications for precision agriculture.
Physical Weed Control in Ontario
Shaun discussed the importance of physical weed control in Ontario, highlighting the need for an integrated approach to manage weeds effectively. He explained the concept of ecology in weed control, emphasizing the understanding of past and present weed ecology to manage them. Shaun also discussed various methods of physical weed control, including tillage, mowing, mulching, and AI considerations. He presented a study on the use of mulch in managing kochia, a problematic weed in the prairies. Shaun also touched on the use of harvest weed seed control, a new paradigm developed in Australia, to prevent weed seeds from going back into the environment. He concluded by expressing his willingness to answer any questions.
Weed Detection and Management Challenges
Shaun presented on weed detection and management, highlighting the challenges of species identification and the need for comprehensive models. He suggested that cloud computing could be used for image analysis, but edge computing might be more suitable for real-time applications. Bridget asked about the next steps for innovation, and Shaun emphasized the importance of publicly available data sets for model training. Ian inquired about similar work in non-field areas, and Shaun acknowledged the potential but lack of specific knowledge. Bridget proposed the idea of a publicly available data set for species-level weed identification, which Shaun agreed could be feasible with collaboration and data storage solutions. Medhat asked about the economic viability of different weed management solutions, and Shaun acknowledged the need for more studies in this area.
Agenda Changes and Upcoming Events
In the meeting, Dan led the discussion, thanking Christopher for his presentation and encouraging others to present in the future. Bridget announced changes to the agendas sent out on Wednesdays, and Taylor reported on the Agrotech Venture Forum. Brian invited companies to showcase their technology at the demo day, and Bridget shared information about the mentorship intake for early-stage robotics companies. Ian announced the Southwest Food Innovation Summit. The conversation ended with thanks to Shawn for his presentation and to all attendees.
