October 10, 2025. Ezeh Ifesinachi, University of Guelph
Meeting summary
Quick recap
The working group meeting welcomed new members and discussed agricultural robotics research, with a presentation on the NIO Oreo robot's weed management capabilities in different soil types. Technical challenges and limitations were explored during the discussion of various robotic systems, including issues with GPS connectivity, mechanical constraints, and weed identification. The conversation ended with updates on upcoming events, research proposals, and initiatives, while also addressing the need for improved validation processes and documentation for agricultural technology.
Next steps
Summary
Agricultural Workforce and Innovation Updates
The working group meeting welcomed new members Phyllis McCallum from the Canadian Agricultural Human Resource Council and David Reed from Waterloo Economic Development Corporation. Phyllis explained her role in workforce strategy and labor market information, while David shared his interest in supporting agricultural businesses and youth innovation in the Waterloo region. The group then heard from Ifesinachi Nelson, a master's student at the University of Guelph, who was scheduled to present on the NIO Oreo robot's weed management assessment in muck and mineral soils, but the presentation was not completed in the provided transcript.
NIO Robot Weed Management Trial
Ifesinachi presented a trial comparing the NIO Robot's weed management efficiency in Holland Marsh with a tractor. The NIO Robot, a fully electric autonomous vehicle, was evaluated for its performance in managing weeds in both mock and mineral soil. The trial aimed to address labor and time constraints faced by growers in the region, with the robot demonstrating reliable operation and features such as RTK and camera-guided navigation, as well as obstacle detection and mapping capabilities.
Robotic vs. Tractor Farming Trial
Ifesinachi discussed the setup and results of a field trial comparing robotic and tractor-based farming methods across two soil types and two crops. The trial used different cultivators and spraying equipment, with some initial delays in implementing the rolling cultivator on the robot due to configuration issues. The robot and tractor showed similar uptime and downtime metrics, with charging being the main cause of robot downtime, and network signal disruptions causing brief interruptions in 2024. While there were differences in weed biomass and beet yields in muck soil, no significant differences were observed in mineral soil between robotic and tractor operations.
2025 Season Technical Challenges
Ifesinachi discussed the challenges faced during the 2025 season, including technical issues with the audio and robot that hindered the repetition of a 2024 study. Despite efforts to compare treated and untreated portions of the Cardiff field, the robot's precision cones and herbicide damage to crops led to the trial's termination. Subsequently, Ifesinachi explored the use of a pixel farming laser machine, which struggled to acquire targets and accurately deploy lasers in the field, resulting in inconsistent and unreliable results. Both trials were terminated early due to these persistent issues.
Oreo vs. Tractor System Trial Results
The team presented findings from their 2-year trial comparing the Oreo and tractor systems, noting that while the Oreo produced reliable results, it faced technical challenges including GPS connectivity issues and mechanical limitations. They identified several next steps including improving GPS reliability, adding better insulation for cables, and developing a more adaptable hitch system. The discussion revealed that the LaserOne prototype could potentially be installed on the Oreo for autonomous driving, though this was not pursued due to mechanical limitations with ground speed and motor controllers. The team encountered various weed species including redroot pigweed, tisels, and lambsquarters across different soil types, and concluded that a full season of testing would be needed to validate findings and measure ROI.
Robot Repair Troubleshooting Challenges
Chuck Baresich discussed a technical issue with a robot where three computers were damaged due to a faulty wiring connection that caused a short circuit. NIO, the robot's manufacturer, initially suggested replacing the computers without conducting proper troubleshooting, leading to a delay of about two months before the root cause was identified. Chuck noted that while NIO provided some support, there was a disconnect between their approach and the user's needs, highlighting the importance of following a systematic troubleshooting process. He emphasized the need for established checklists and procedures, which NIO lacked at the time, and mentioned that documenting repairs would help create such resources for future users.
Agricultural Robot Validation Discussion
Chuck explained that he deliberately chose not to send an Oreo robot to a farmer for spring use due to concerns about its readiness, and similarly, he was not convinced that the Pixel 1 laser weeder was fully ready for commercial use, opting instead for a small-scale trial. Ian and Bridget discussed the challenges of weed identification and targeting in robotic systems, with Ian suggesting the use of pre-trained weed ID programs and Chuck noting the complexity of distinguishing crops at different growth stages and the need for precise laser application. Bridget inquired about the availability of a validation checklist for new agricultural technology, and Chuck expressed openness to expanding and continuing this effort, particularly for the Oz robots, with the possibility of including other technologies in the future.
Research Proposals and Outreach Updates
The meeting covered several updates and discussions. Bridget announced the opening of research proposals for the next growing season, with funding available through the AgRobotics Working Group for projects starting in April 2026. She also mentioned ongoing funding for current-year projects and encouraged participation in various subcommittees, including research, education, demo day, and Ag Validation. Chuck shared excitement about the first high school outreach for the robotics competition, highlighting interest from both small-town and larger schools. The conversation ended with an open roundtable discussion, inviting participants to voice questions and comments.
AgTech Events and Robotics Updates
Ian discussed his upcoming presentation at the Ag Conference, highlighting three examples of robotic units in Ontario fields. Joe mentioned upcoming Ag Enlightened and Garson events in Winnipeg on October 22nd and 23rd. Ainsley reported on the Canadian Greenhouse Conference, noting increased focus on automation and robotics. Bridget proposed dates for a one-day AgTech and AgRobotic startup event, with January 16th emerging as the preferred date after considering potential conflicts.
Strawberry Laser Project and Conferences
The group discussed several upcoming events and initiatives. Konrad presented a project on using laser technology to cut strawberry runners, seeking input on identification algorithms and laser implementation. Dan mentioned the Canadian Weed Science Society's annual meeting in November and various conferences, including the Robotics and AI Symposium in Kingston. Debjani requested connections with conservation companies for her groundwater level monitoring technology. The Ontario Pest Management Conference was briefly mentioned, having found a replacement speaker for a cancelled presentation. The conversation ended with a reminder about the upcoming Thanksgiving long weekend and the next meeting on the 17th.
