UC Berkeley team advances food systems in new $20M AI research center

October 01, 2020

A new research center funded by the National Science Foundation (NSF) in partnership with the U.S. Department of Agriculture (USDA) and the National Institute of Food and Agriculture (NIFA) will aim to improve U.S. food systems to address such issues as pandemic-driven food system security and safety; improving crop yield, quality and nutrition; decreasing energy and water resource consumption; and increasing production and eliminating food waste.

The NSF awarded $20M over five years to create the USDA-NIFA AI Institute for Next Generation Food Systems (AIFS), one of five AI institutes established to accelerate research and support the U.S. workforce. The center is led by a team at UC Davis in partnership with researchers from UC Berkeley, Cornell University, University of Illinois at Urbana-Champaign, the UC Division of Agricultural and Natural Resources (ANR) and the U.S. Department of Agriculture. Each of these institutions has a proven track record of excellence across such areas as artificial intelligence (AI), simulation, food system sciences and engineering.

Professor of mechanical engineering Tarek Zohdi will lead the UC Berkeley team, which is comprised of members of the Center for Next Generation Food Systems and includes Ethan Ligon, a professor in the Department of Agricultural and Resource Economics. Together the Berkeley team offers collective expertise in high-performance computing, advanced manufacturing, pathogen propagation, contact tracing, decontamination, autonomous systems and technology transfer.

Ligon commented that this funding creates an opportunity to bring the world's leading researchers in agricultural & resource economics in closer contact with world leading AI researchers. "My research shows that contracts in agriculture often depend on details of quality measurement and other environmental factors in a complicated way,” he said. “I'm really excited to be working with world leading AI researchers to bring machine learning tools to bear on the problem of better understanding these complicated contracts, and perhaps designing better ones."

“Berkeley has extensive expertise in simulation technologies, which can be applied to create virtual food systems that will leverage the monumental leaps in high-performance computer simulation, AI and machine learning,” said Zohdi. “This new effort will be centered around the concept of the ‘Digital Twins’ of physical reality — digital replicas of complex food systems that can then be inexpensively and safely manipulated, improved and optimized in a virtual setting.”

Read the original post by Berkeley Engineering.