U-M leads research translating semiconductor innovation for broad societal impact

Video by Eric Shaw

[email protected]

Semiconductors play a critical role in the fabrication of electronic devices, leading to important advancements in areas ranging from clean energy and health care to military defense systems. The University of Michigan is at the forefront of semiconductor research, partnering with government and industry to advance innovative discoveries and technologies for broad societal impact.

At the University of Michigan, about 100 faculty representing a variety of disciplines – from engineering to public policy – are working closely with government and industry partners to advance innovative semiconductor research.

Article by Kelsey Keeves

[email protected]

David Blaauw

Kensall D. Wise Collegiate Professor of Electrical Engineering and Computer Science

photo of David Blaauw

Professor David Blaauw

David Blaauw, a professor of electrical engineering and computer science at U-M, is a national leader in shaping the future of semiconductors through his involvement in the newly launched Joint University Microelectronics Program 2.0 (JUMP).

He and his team are working to develop a sensor that can more efficiently and accurately perceive its surroundings and only attend to the most pertinent information.

They will do this by creating a digital feature processing engine that only refines the part of the image that has something happening within it. Helping the sensors automatically narrow the scope of what data needs to be processed allows the entire system to use less power and therefore be more efficient.

Blaauw joined U-M in 2001 as an assistant professor in the Department of Electrical Engineering and Computer Science after working as an engineer at IBM and Motorola.

With assistance from U-M units like Innovation Partnerships, which helps faculty commercialize research for broad societal impact, he has been able to translate several of his research projects into real-world applications. His research has led to 73 patents and several startup companies, in areas from hardware acceleration for image processing to low power and adaptive design.

“When you actually see the technology going into the market field, that puts a whole other perspective on it, which then not only impacts society, but also creates jobs and benefit communities,” said Blaauw.

“Taking technologies through that path actually also circles back to raise new questions that otherwise, you wouldn’t think about, and because of this, it opens new research problems and new research topics.”

Blaauw hopes to continue to apply his work to interdisciplinary projects in the medical field and biological sciences, tackling the problems that fall between traditional fields of work. His past work has assisted his U-M colleagues in helping to track the migration path of monarch butterflies and using the world’s smallest complete computers to find out why indigenous snails went extinct in Tahiti.

“We are seeing some of the quickest development of new technology in this field that’s both very challenging, but also provides lots of opportunities. We really can break barriers and make an impact on society.”

Valeria Bertacco

Arthur F. Thurnau Professor
Mary Lou Dorf Collegiate Professor of Computer Science and Engineering
Vice Provost for Engaged Learning, Office of the Provost
Professor, Electrical Engineering and Computer Science

photo of Valeria Bertacco

Professor Valeria Bertacco

Valeria Bertacco is an Arthur F. Thurnau Professor, the Mary Lou Dorf Collegiate Professor of Computer Science and Engineering and a professor of electrical engineering and computer science.

She also serves as the Vice Provost for Engaged Learning, advocating for university initiatives that create opportunities for action-based student learning and supporting international partnerships for the institution.

Bertacco began her career working for a company that designed software for computer chips. After four years, she realized she missed the collaboration that happens in university research environments.

She joined the University of Michigan in 2003 as an assistant professor of computer science and engineering after receiving her Ph.D. in electrical engineering from Stanford University.

“I have found that my industry experience was very, very useful when I became a professor because I could teach the theory of logic design to my students and connect it to experiences I had.”

One of the reasons Bertacco chose to join U-M was because of the support and community she found while on campus. Collaborating with students and fellow researchers is what she values most about her role as a U-M faculty member and researcher.

“The most energizing part of my work is in doing research and brainstorming with the students.

“I love the technical interactions. All my ideas come from discussions with colleagues and students.”

Her research focuses on the design of robust and efficient computing systems, with her most recent project on accelerating explainable AI, a framework that helps us understand the reasoning behind the decisions and predictions made by AI, to increase its practical use.

Bertacco is also director of the U-M based Applications Driving Architectures Center, which aims to enable increased participation in hardware design through lower cost manufacturing and easily adoptable design infrastructures.

In support of this role, and encouraged by College of Engineering leadership, she founded the Michigan’s Advanced Vision for Education and Research in Integrated Circuits, or MAVERIC, a new microelectronics research and education initiative that aims to leverage the collaboration and expertise at U-M to position Michigan as a leading hub in the United States’ mission to reshore semiconductor manufacturing.

“It is so rewarding to witness the growing focus around semiconductor innovation and translational research. I believe that over the next few years, we will witness major growth in this discipline, and I know that Michigan faculty will have a big role in this expansion, especially in the space of automotive semiconductors.”

Wei Lu

Professor of Electrical Engineering and Computer Science

headshot of Wei Lu

Professor Wei Lu

Wei Lu, a professor of electrical engineering and computer science, is leading the way in creating powerful and efficient AI systems, and was named the Distinguished University Innovator of the Year in 2022 for his pioneering efforts in the development and commercialization of novel electrical devices.

Lu joined the U-M Electrical Engineering and Computer Science Department in 2005 after completing a postdoctoral research fellowship at Harvard University. His undergraduate degree was focused on physics, which he believes gives him a unique perspective on the limitations and challenges faced in creating advanced electronics.

When he first began studying engineering and technology, the field was focused on how to make devices smaller. But now that many components have evolved to be as small as physics allows, the field is focused on other areas that were previously overlooked.

Lu believes that the challenges presented by the limitations of the technology provide opportunities for new innovation.

His startup, MemryX, has developed a programmable AI accelerator chip with unmatched performance and relatively low power.

MemryX is located in the Innovation Partnerships’ Startup Incubator, which provides resources to qualified U-M startups that are emerging from the pipeline of Ventures startup projects. The Incubator offers flexible leases for world-class laboratory and office space in Michigan’s North Campus Research Complex. Tenants are provided with concierge service and networking opportunities to accelerate their programs guided by the Ventures team.

“Because the problem we started with doesn’t have a known solution, it forces you to be creative and have a design that can lead to something really exciting, and like with MemryX, potentially commercialize it as well,” said Lu.

Lu came to U-M because of the interdisciplinary and collaborative structure that U-M has, as well as its supportive network of students and researchers. He emphasized the importance of cross-disciplinary collaboration as the key to encouraging new possibilities and areas of research, as it allows researchers to search for problems at the edges of the discipline and encourages new approaches that might have existed within the field without researchers coming at it from a different field and perspective.

“The bottleneck in computing today is not due to computing itself – it is really due to the cost of moving data muscle memory. So we started looking at devices with memory-centric computing architectures. That actually helped us break into the computing field and this gives us a fresh perspective because, normally, when building a new computing architecture, most follow an existing path.”