Innovative Agricultural Robotics: A Solution to Superweed Challenges

Most corn and soybean fields in the U.S. rely on herbicide-resistant crop varieties. However, the emergence of superweeds resistant to common herbicides threatens current weed management strategies. A new study from the University of Illinois Urbana-Champaign explores how different types of farmers and fields are likely to adopt agricultural robotics for mechanical weeding and at what stage of resistance development.

Madhu Khanna, a professor of agricultural and consumer economics in the College of Agricultural, Consumer and Environmental Sciences (ACES) and director of the Institute for Sustainability, Energy and Environment at Illinois, explained, "The exclusive reliance on herbicides for weed control has led to the appearance of superweeds, and we don't have anything in the pipeline in terms of new modes of action. If chemical control methods fail, it could result in millions of dollars per year in crop losses."

Small, lightweight robots designed to operate under crop canopies offer an efficient and environmentally friendly alternative. These robots use artificial intelligence for navigation and automate the process of pulling hoes through the soil to disrupt the emergence of weed seeds. Although not yet commercially available for corn and soybean crops, they hold promise for managing persistent threats like waterhemp (Amaranthus tuberculatus), which has already developed resistance to multiple herbicides.

The researchers investigated two distinct weed management strategies: myopic management, which considers immediate consequences, and forward-looking management, which anticipates future impacts. They assessed factors such as weed seed density, resistance levels, and economic thresholds that might trigger the adoption of robotic weeding at the farm level.

According to co-author Shadi Atallah, an associate professor in ACE, "We found that both seed density and resistance level are important for myopic management. For a forward-looking approach, seed density does not matter because resistant seeds are likely to spread in the future. This perspective considers resistance levels, but almost any level can trigger adoption."

If a robot costs around $20,000, farmers with a forward-looking mindset are likely to adopt it when only 0.0001% of the seeds are resistant. In contrast, those with a myopic approach will wait until resistance levels exceed 5%. Atallah emphasized, "If you're managing for the future, don’t even bother to look at seed density; just look at the resistance level, and no matter how low that is, you should go ahead and adopt the robots."

The study also evaluated the adoption rate and intensity over time. Findings indicated that farmers with a myopic perspective would refrain from using robots in the first six years, continuing to apply herbicides until their effectiveness waned before shifting entirely to robotic control in year seven. Conversely, forward-looking farmers would adopt robots sooner, using them to complement herbicide treatments to preserve their efficacy.

At the outset, myopic management may lead to higher profits due to lower initial investments, but forward-looking management yields better long-term returns. Atallah pointed out, "Farmers may adopt a myopic perspective if they lease land annually, preventing long-term planning. However, even those managing on a yearly basis will eventually need to adopt robots as other control options become ineffective."

These differing management strategies have implications beyond individual farms, as resistant seeds can migrate to neighboring fields. A forward-looking approach can help reduce resistant seed populations, potentially reversing resistance trends. While resistance isn’t reversible for all weed species, in the case of waterhemp, the reproduction rate of resistant seeds decreases under lower selection pressure, allowing non-resistant seeds to prevail over time.

While this research focused on maximizing profit at the farm level, future studies will explore the spillover effects of resistant seeds on neighboring farms and evaluate landscape-level impacts, providing further insights for policymakers. Atallah presented the findings, along with results from a farmer survey, during a farmdoc daily webinar. This research was funded by AIFARMS, an AI Institute at the University of Illinois, supported by USDA's National Institute for Food and Agriculture.