A team headed by Purdue University has been selected for a project to enhance artificial intelligence hardware through the Microelectronics Commons program in partnership with the Silicon Crossroads Microelectronics Commons Hub.
Purdue said Monday its CHEETA: CMOS+MRAM Hardware for Energy-EfficienT AI project, one of four awardees announced on Sept. 18, aims to make AI hardware more energy efficient and the sensor-to-decision response time faster. This development will enhance the “size, weight and power” metrics, which are expected to support defense initiatives.
The project intends to utilize CMOS+X and magnetic random-access memory, or MRAM, to develop efficient in-memory computing hardware fabrics.
The team will receive $21 million for over four years from the Applied Research Institute to fund the project. It is made up of faculty members from the Elmore Family School of Electrical and Computer Engineering and collaborating partners Northrop Grumman, Everspin Technologies, Naval Surface Warfare Center Crane Division, Argonne National Laboratory, Georgia Institute of Technology and University of Michigan.
Purdue President Mung Chiang, stated, “Purdue is strongly committed to supporting the CHIPS Act and the U.S. DOD’s ME Commons program, enhancing economic prosperity to the Silicon Heartland and national security of the U.S.”
Karen Plaut, executive vice president for research at Purdue, emphasized that the collaboration will enhance defense technologies and boost the nation’s leadership in advancing critical technologies.
“By uniting the strengths of our partnerships with Argonne National labs, NSWC Crane as well as government agencies, leading industry partners and top academic institutions, we are embarking on a groundbreaking project that will push the boundaries of semiconductor innovation,” Plaut said.