Washington State University will lead a new federally funded research institute to take the agriculture industry further into the future via artificial intelligence.
The USDA-NIFA Institute for Agricultural AI for Transforming Workforce and Decision Support — also known as the AgAID Institute — will look at how AI can help tackle farming challenges related to climate change, weather, water supply and labor.
To that end, WSU researchers will lead a coalition of colleges and universities as well as private-sector partners through the AgAID Institute, which will be funded through a five-year, $20 million U.S. Department of Agriculture grant.
“We also know that agriculture has entered in some ways a new era. It’s what some people talk about as Agriculture 4.0,” said WSU professor Ananth Kalyanaraman, director of the AgAID Institute. “It’s almost an evolution that is needed in agriculture to tackle some of the 21st-century challenges.”
AgAID is one of 11 new AI research institutes announced Thursday by the National Science Foundation.
The University of Washington will lead another: The NSF AI Institute for Dynamic Systems will focus on AI innovations with real-time learning and control of complex dynamic systems, which describe chaotic situations caused by constantly shifting conditions, according to the university.
“Some of our specific questions include: Can we develop better machine-learning technologies by baking in and enforcing known physics, such as conservation laws, symmetries, etc.?” institute associate director Steve Brunton, a UW associate professor of mechanical engineering, said in a statement. “Similarly, in complex systems where we only have partially known or unknown physics — such as neuroscience or epidemiology — can we use machine learning to learn the ‘physics’ of these systems?”
Kalyanaraman, a professor with WSU’s School of Electrical Engineering and Computer Science, will lead the AgAID Institute.
Partners with the institute include Oregon State University; the University of California, Merced; the University of Virginia; Carnegie Mellon University; Heritage University; Wenatchee Valley College; Kansas State University; IBM Research; and the start-up innov8.ag.
Kalyanaraman said WSU first formed an internal team of computer science researchers and agricultural stakeholders a few years ago to look into the perceptions and practical applications of AI with farming. The team then grew with external entities, including a number of the institute’s partnering universities.
These partnerships incidentally put the coalition in prime position when the National Science Foundation put out a call for proposals with the research institute initiative last year, Kalyanaraman said.
One of the focuses of the institute will be education programs from K-12 through higher education and worker training. Kalyanaraman also said much of the institute’s initial years will be spent working directly with stakeholders in the agricultural industry — essentially the end users of these systems — to determine their needs and practical restraints.
“It’s one thing to have AI tools and design AI tools to work in lab settings,” he said, “but it’s a completely different beast when we talk about transferring AI to work while in the field.”
AI, Kalyanaraman said, can help farm managers better forecast the availability of water and allocate that supply in smart ways to prevent significant losses. He said such innovations can also help farmers better predict the weather and plan for extreme weather.
“How can you take the data that is being generated and convert that into some kind of actionable knowledge to make real-time decisions?” Kalyanaraman said. “That is where AI comes in.”
Labor, meanwhile, is another challenge — particularly with specialty crops such as apples, cherries, mint and almonds that require skill-intensive processes. Specialty crops account for 87 percent of the U.S. agricultural workforce, according to the university.
As an example, Kalyanaraman said an AI-driven device could guide a farm worker in pruning crops with adroitness.
But don’t expect robots to manage these specialty crops anytime soon.
“A machine simply can’t substitute the human on the farm,” Kalyanaraman said. “Farmworkers are an integral part of the chain, and they will continue to be. Except now with machines, we can help train them better and also have some degree of automation in the sense that guidance is possible.”
U.S. Sens. Maria Cantwell and Patty Murray, both Washington Democrats, lauded Thursday’s announcement.