Artificial Intelligence Technologies for Agricultural Sustainability
Market Corner by Madhu Khanna, C-FARE Board Member and Professor at University of Illinois, Urbana-Champaign
Productivity growth in the agricultural sector has benefited from numerous technological innovations, including the development of high yielding varieties that were more responsive to fertilizer, introduction of genetically modified crops that made crops resistant to pests and, more recently, digital technologies that can map spatial variability in growing conditions within a field and vary input applications in response to it. The advent of artificial intelligence (AI) applications in agriculture is emerging as the next wave of innovation that can fundamentally change the way that agricultural decision making and operations are conducted. These AI technologies have the potential to address three critical challenges for agriculture. The first is the need to manage the collection and analysis of vast amounts of spatial data about growing conditions, plant growth, disease conditions, and nutrient deficiencies within a field. The second is to address the growing shortage of labor for farm operations by developing autonomous equipment that can rely on intelligent machines and robotics to make site-specific interventions in the field in ways that are less labor intensive. The third is the need to achieve continued growth in productivity while reducing environmental harm to soil and water quality and regenerating soil health.
Specifically, small robots that can operate under the canopy of row crops like corn and soybeans to plant cover crops and mechanically remove weeds are promising alternatives to conventional technologies. By planting cover crops under the canopy of a standing crop, they can enable better establishment of the cover crop early in the growing season and lower the labor intensity of the operation. This can enhance the soil carbon sequestration and soil health benefits of cover crops. Robots that mechanically remove weeds provide a mechanism for reducing damages to crop yields from herbicide resistant weeds and adverse impacts on the environment from chemicals.
These technologies are scale neutral and therefore could be adopted by small and large farmers. While benefits may materialize in the longer term and a large component of these benefits may be public rather than private, upfront costs of adoption for farmers can deter adoption. Other factors that will affect adoption include access to high-speed broadband, availability of technical assistance and mechanisms for reducing the risks of adoption. Policies that value the public benefits to soil and water quality can also induce adoption of AI technologies that improve agricultural sustainability.
