Cleaveland, Marcus part of $10 million NSF collaborative research grant

Professor Rance Cleaveland (CS/ISR) is the principal investigator and Professor Steve Marcus (ECE/ISR) is a co-PI for the University of Maryland?s portion of a major new National Science Foundation collaborative research grant, ?Next-Generation Model Checking and Abstract Interpretation with a Focus on Embedded Control and Systems Biology.? The five-year, $10 million project is part of NSF's "Expeditions in Computing" initiative.

Maryland's part of the project is worth $1.8 million. Along with Marcus, Tongtong Wu of the University of Maryland?s School of Public Health is also a co-PI. Wu is an assistant professor of biostatistics in the Department of Epidemiology and Biostatistics.

The consortium will develop new computational tools to help scientists and engineers analyze and understand the behavior of the complex models they develop for application domains ranging from systems biology to embedded control. Building on the success of model checking and abstract interpretation (MCAI), two well-established methods for automatically verifying properties of digital circuit designs and embedded software, this research project will extend the MCAI paradigm to systems with complex continuous dynamics and probabilistic behaviors.

What is MCAI?
Model checking and abstract interpretation are the result of more than 30 years of research. Model checking is the most widely used technique for detecting and diagnosing errors in complex hardware and software designs. It considers every possible state of a hardware or software design and determines if it is consistent with the designerÂ’s specifications; it produces counterexamples when it uncovers inconsistencies. It is limited, however, by the size of the systems it can analyze.

Abstract interpretation, by contrast, doesnÂ’t attempt to look at every possible state of a system, but to develop a simplified approximation of a system that preserves the particular properties that need to be assessed. This makes it possible to analyze very large, complex systems, such as the one million lines of code in the Airbus A380Â’s primary flight control system, but with less precision than is possible with Model Checking.

In this new project, the researchers plan to take advantage of the strengths of both methods by tightly integrating the two into what they call MCAI 2.0.

Scope of the research
The research will include: understanding the precursors and course of pancreatic cancer; predicting the onset of atrial fibrillation; and obtaining deep design-time insights into the behavior of automotive and aerospace control systems. Ultimately, the project is expected to provide vital tools that will enable health care researchers to discover better treatments for disease and will allow engineers to build safer aircraft and other complex systems.

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 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,? Cleaveland says.

Cleaveland works on embedded software, such as that used in cars for antilock brakes or for flight control on planes. His research 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. 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.

Carnegie Mellon University is the lead institution for the program. In addition to the University of Maryland, other collaborating institutions are the City University of New York, New York University, SUNY Stony Brook, Cornell University and NASA?s Jet Propulsion Laboratory.

The world-class team of scientists and engineers assembled for this Expedition includes two Turing Award winners, a recipient of the National Medal of Science, and awardees of other prestigious research prizes.

| Website for the project | Read a National Science Foundation story about the project |

Published August 14, 2009