AMES, Iowa – The latest artificial intelligence tools will allow researchers to develop digital twins of individual crop plants and entire farm fields, helping plant breeders improve crop varieties and farmers boost production.
The researchers behind a new artificial intelligence research institute say their work can accelerate the productivity and sustainability of agriculture at a time when the world’s population is increasing, cropland is decreasing and the climate is changing.
The National Science Foundation and the U.S. Department of Agriculture’s National Institute of Food and Agriculture are supporting the researchers’ idea with a five-year, $20 million grant to establish an AI Institute for Resilient Agriculture (AIIRA – “eye-rah”) based at Iowa State University. The institute is one of 11 AI institutes announced today.
“I am delighted to announce the establishment of new NSF National AI Research Institutes as we look to expand into all 50 states,” said National Science Foundation Director Sethuraman Panchanathan. “These institutes are hubs for academia, industry and government to accelerate discovery and innovation in AI. Inspiring talent and ideas everywhere in this important area will lead to new capabilities that improve our lives from medicine to entertainment to transportation and cybersecurity and position us in the vanguard of competitiveness and prosperity.”
The new institutes join seven previously announced in August 2020. They’re all part of a $360 million ($220 million this year and $140 million last year) federal effort to develop hubs for artificial intelligence research that address national needs.
“AIIRA is bringing together scientists and farmers, industry and government to adapt these technologies and encourage their adoption to more effectively transform agriculture to meet the needs of our growing and increasingly climate-challenged planet,” project leaders wrote in a summary.
Baskar Ganapathysubramanian, the Joseph C. and Elizabeth A. Anderlik Professor in Engineering at Iowa State, will lead the institute. It will include collaborators from Iowa State, Carnegie Mellon University, the New York University Tandon School of Engineering, the University of Arizona, the University of Nebraska-Lincoln, George Mason University, the University of Missouri and the Iowa Soybean Association.
The institute also includes collaborators from the tech and agriculture industries, governments, commodity groups and other organizations.
“These are problems that can’t be answered by any individual,” Ganapathysubramanian said. “We need engineers, data scientists, plant scientists, social scientists, farmers, educators and entrepreneurs. AIIRA will bring all this expertise together.”
The new institute’s university base is already home to a lot of that expertise, said Iowa State President Wendy Wintersteen.
“AIIRA is perfectly suited for Iowa State University as it leverages our core strengths in the plant sciences, engineering and technology and our culture of collaborative innovation,” she said. “Together with our partner institutions, Iowa State will propel agricultural advancements that will have a powerful impact on farmers, entrepreneurs, and industry in Iowa and across the country.”
Carrie Castille, director of the USDA’s NIFA, said the new institutes will “leverage the scientific power of U.S. land-grant universities” such as Iowa State. And, she said, “These innovation centers will speed our ability to meet the critical needs in the future agricultural workforce, providing equitable and fair market access, increasing nutrition security and providing tools for climate-smart agriculture.”
The big idea – “predictive digital twins”
An illustration of the AIIRA idea – headlined “Deploy the twin” – shows two research scenarios:
1.) The breeding-scale scenario pictures a single plant growing from a lump of brown soil, its five leaves a healthy green. Then there’s an arrow pointing to an identical plant, but this one is an electric blue with highlights showing digital networks. In between is a robotic arm poised to examine a plant.
2.) The production-scale scenario pictures neat rows of green plants on a striped, brown field. The corresponding digital twin is all in shades of blue. In between flies a drone sending data between the real field and the digital one.
The scenarios illustrate how AIIRA researchers will continuously feed real-time data from sensors in the biological plants and fields – including weather data, soil measurements of water and nitrogen, soil and topography maps, ground and drone-based imaging and satellite information – to the predictive digital twins.
In addition, the digital twins will be loaded with the latest understanding of plant growth and development.
“This will be a functional, virtual representation of a real system,” said Soumik Sarkar, the institute’s associate director, an associate professor of mechanical engineering and a Walter W. Wilson Faculty Fellow in Engineering at Iowa State. “With it, we can run various what-if scenarios and make decisions in the virtual world, and when we’re happy with the results, we can apply those decisions in the real world.”
And because every year of biological data can create hundreds of reality-based simulations, a project summary says:
- Scientists will gain insights into plant genetics and physiology while more efficiently breeding plants that can adapt to changing environments and emerging diseases and pests.
- Farmers will optimize planting decisions and applications of fertilizer, irrigation and pesticides to maximize proﬁt while minimizing environmental impacts.
- Governments will identify which crops and practices are best suited on regional and national scales, set policies, and undertake regional and national pest surveillance.
- Ag companies will create more effective precision management products and produce climate-resilient crop varieties.
- And, rural entrepreneurs will be spurred by artificial intelligence-fueled innovations and economic development.
Investments that grow
Some of these applications of artificial intelligence and digital twins have started in the manufacturing and pharmaceutical industries, Sarkar said.
“The big jump is bringing this to the natural or biological world,” he said.
It’s a jump that, over time, could pay big dividends.
“This is a little like an investment portfolio,” said Patrick Schnable, an Iowa State Charles F. Curtiss Distinguished Professor in Agriculture and Life Sciences, the Iowa Corn Promotion Board Endowed Chair in Genetics, the Baker Scholar of Agricultural Entrepreneurship and director of Iowa State’s Plant Sciences Institute.
“When you invest in stocks and bonds, you’re trying to come up with the best investment mix based on personalized financial information,” he said. “With this project, by planting crops virtually without having consequences in the real world, you can find the right mix of plant varieties and management practices to produce the best crops.”
The AI technology hasn’t been around to make that a possibility for agriculture, said Arti Singh, an Iowa State assistant professor of agronomy. Improvements in plant, soil and weather sensors, along with automated devices such as robots and drones now make it possible to collect real-time data and transmit enough of it to build useful digital twins capable of predicting agricultural production processes.
The same goes for predicting the performance of new crop varieties.
“How can you breed for the environment or stressors that will prevail in 2050?” asked Asheesh (Danny) Singh, a professor of agronomy and the Bayer Chair in Soybean Breeding at Iowa State. “AIIRA can help us optimize plants with the right combinations of roots, stems, leaves and reproduction.”
The institute will also educate students, scientists, business people and farmers to understand and use these new digital tools to help them make better decisions. The institute, its leaders wrote, “is committed to speeding the progress, productivity and sustainability of today’s agriculture by making the power of artificial intelligence available to all.”