Nov 9, 2020Artificial intelligence: The next revolution in agriculture
Hidden among all the news about COVID-19, hurricanes and the presidential election, you may have missed one of the most important news stories of the year.
In the center of its campus in Gainesville, the University of Florida is building the most powerful artificial intelligence supercomputer in U.S. higher education. This has been huge news at UF, as it will create exciting new opportunities for faculty, students and researchers. But what does it mean for growers and stakeholders in Florida?
Artificial intelligence is the ability of a computer system to recognize patterns, understand language, learn from experience, solve problems and perform complex tasks. In other words, it’s the ability of a machine to think like a human brain, but to do so faster, more accurately and on a massive scale.
If you’re a grower, you care about every plant and every animal on your farm. Given enough time and manpower at your disposal, you would scrutinize every plant and animal, giving each the exact resources needed to grow to their maximum potential. AI allows you to do just that, putting eyes on millions of variables and coordinating vast amounts of data instantly and with pinpoint accuracy.
AI requires a massive amount of computing power. This is why UF’s partnership with NVIDIA and UF alumnus Chris Malachowsky’s gift of the $70 million supercomputer, named HiPerGator, is such an important tool for agriculture.
To give you an idea of the capacity of UF’s new HiPerGator 3.0 computing system, if every UF student doing homework, every faculty member and scientist conducting research, every facility running logistics and every budget office crunching numbers were all using HiPerGator at the same time, it would only tap about 15 percent of its full capacity.
What can it do with other 85 percent? That’s for all of us to decide.
Here at UF/IFAS we’re excited about the possibilities of applying AI to solve everyday problems related to supporting Florida’s agriculture industry, feeding the world’s growing population and using natural resources sustainably. AI has the promise to significantly advance our ability to strengthen food chain systems, develop precision breeding of plants and animals, apply robotics in food and agriculture operations, track the location and spread of pests and pathogens, and collect data on agroecosystem services.
In fact, UF/IFAS has already been applying AI to improved agricultural production for many years. HiPerGator will serve to super-boost these efforts and bring new resources to grow AI’s abilities to help Florida’s growers be more productive and save resources. The following are just a few of the areas where UF/IFAS Extension is helping Florida’s growers apply AI research to their production.
In the animal sciences department, Albert De Vries’ team is using AI to get more accurate profiles of individual cattle, including measuring their phenotypes to aid in breeding, monitoring their feed intakes during group feeding, and using their genetic makeup to improve feeding efficiency.
Yiannis Ampatzidis and his research team at Southwest Florida REC have developed AI-based software called Agroview to analyze and visualize data collected from UAVs, or drones. Instead of spot checks, UAVs can take images of thousands of individual plants and upload them to cloud-based software that analyzes the data to access plant qualities, quantities and growth factors or impacts. The software has a wide range of applications for Florida’s agriculture, including our iconic citrus industry. Developed to help producers take better care of their crops while also saving money, Agroview is such a revolutionary innovation that it recently won UF’s invention of the year award.
Determining the maturity of peanut seed currently requires shelling the hulls from test samples and making a subjective estimate based on seed color. Now Agronomy Department Chair Diane Rowland and her research team have developed a method using hyperspectral imaging and AI to determine peanut seed quality through the hull, allowing peanut farmers to select mature seeds with far greater accuracy and less expenditure of time and labor.
Daniel Lee, Feng Wu, Abd-Elrahman, Natalia Peres, and Shinsuke Agehara are using drones and tractor mounted cameras to take images of strawberry flowers and fruit, compiling the data to predict where and when harvesting should occur to optimize yield and resources.
At a new facility at the Gulf Coast REC, weed scientist Nathan Boyd and precision agriculture specialist Arnold Schumann are using AI to identify weeds in the field and distinguish them from crops so that herbicides can be applied only to the weeds, resulting in less spray damaged plants and reduced pesticide use.
These are just a few of the research and Extension projects that are already underway at UF/IFAS. UF’s AI initiative kicks into high gear when HiPerGator 3.0 is fully online in January 2021, and the number of the number of AI applications is expected to increase rapidly.
Florida’s growers can anticipate a new era of what Dr. Yiannis Ampatzidis calls the “fourth revolution in agriculture”– applying precision agriculture, robotics, and artificial intelligence to transform the way we feed the world and grow our economy.
For more information, you can view an October 19 Extension Connections conversation with Dean for Research Rob Gilbert and Chair of the Agricultural and Biological Engineering Department Kati Migliaccio about Artificial Intelligence.
Artificial Intelligence at the University of Florida is the place to learn more about how AI is transforming research, teaching and outreach at the University of Florida.
Blogs.IFAS is an excellent source of up-to-the-minute information about UF/IFAS’ latest research and Extension work using Artificial Intelligence.