UM Part of $10 Million NSF Grant for Computational Modeling and Analysis

By Tom Ventsias and Lauren Brown

COLLEGE PARK, Md. - The National Science Foundation has awarded $1.8 million to three University of Maryland professors to develop computational techniques for analyzing models for such disparate applications as aircraft safety systems and pancreatic cancer.

The award, announced last week (Aug. 11), is part of a five-year, $10 million grant in NSF's "Expeditions in Computing" initiative. Maryland faculty are teamed with scientists at Carnegie Mellon University, the lead institution for the program, along with researchers at the City University of New York, New York University, SUNY Stony Brook, Cornell University and NASA's Jet Propulsion Laboratory.

Computer science Professor Rance Cleaveland (far left) is the principal investigator of Maryland's role in the project; Steve Marcus (left), a professor of electrical and computer engineering, and Tongtong Wu (right), an assistant professor of biostatistics in the School of Public Health, are co-investigators. Cleaveland and Marcus have joint appointments in the A. James Clark School of Engineering's Institute for Systems Research.

Cleaveland says he, Marcus and Wu hope to develop revolutionary techniques for automatically analyzing and predicting the behavior of biological and control systems. Using the new techniques, scientists and engineers will be able to greatly accelerate the pace of their discoveries by automating tasks that currently must be performed manually.

The Maryland team is looking to build upon scientific methods developed by the computer science and electrical engineering communities for determining if software and computer hardware behave correctly. The UM research will extend the capabilities of these testing methods from digital systems to more complex systems arising at the intersection of computing and the natural world.

Cleaveland notes that the work is intrinsically multidisciplinary. "Up to now, our team has been working independently -- this brings together the three of us on new work that draws on all our backgrounds," he says.

Cleaveland works on embedded software, such as that used in cars for antilock brakes or for flight control on planes. His research funded by the NSF grant will seek to better understand the behavior of automotive and aerospace control systems.

Wu, whose research interests include computational statistics and statistical genetics, will focus on cancer classification and the genetic determination of diseases, including predicting the onset of pancreatic cancer and atrial fibrillation. She says this research may help identify new ways of verifying models of biological systems with spatial (e.g. gene networks) and temporal (e.g. cell cycles) effects.

Marcus studies control and systems theories and will be developing mathematical models that take into account uncertainties in the systems studied by Cleaveland and Wu. He will also study how to model the composition of such systems.

The overall team of 19 scientists and engineers assembled for this NSF multi-institutional project includes two Turing Award winners, a recipient of the National Medal of Science and awardees of other prestigious research prizes.

The $10 million in NSF funding includes developing a highly ambitious cross-disciplinary educational program called "Complex Systems Science Engineering" and an annual minority-focused workshop for undergraduates on understanding complex embedded and biological systems.

While these programs are held at other institutions involved with the NSF grant, University of Maryland students are eligible to participate, and there are other research opportunities available for Maryland graduate and postdoctoral students within the NASA JPL Research Affiliates Program.

09131