Energy-efficient computers optimally designed for custom applications. New tools to make air travel safer and healthcare interventions more effective. Robotic ‘bees’ that lend a helping hand in search and rescue operations. All of these are promised by new advances in computer technology.
As part of the Expeditions in Computing in August of this year. The National Science Foundation funded several projects at $2 million per year for five years, these projects represent some of the largest single investments made by the directorate.
“NSF supports Expeditions in Computing to stimulate and leverage the tremendous creativity of the computing research community,” said Deborah Crawford, deputy assistant director for CISE. “The projects we support allow academic researchers and their collaborators to explore ideas that promise significant advances in our understanding of the computing frontier, while also yielding great benefit to society.”
The first Expeditions in Computing were established in 2008. With funding appropriated to NSF through the American Recovery and Reinvestment Act (ARRA), the agency is supporting three new trailblazer Expeditions in 2009, bringing the number of projects currently supported to seven.
The three Expeditions awarded in 2009 address a diverse set of challenging problems in computing. “This year’s Expeditions projects are driven by challenge problems that arise from the most pressing issues facing our society today–groundbreaking research shaped by societal needs,” said Mitra Basu, program officer for the Expeditions program.
The awards each feature a lead principal investigator working in conjunction with collaborators at multiple institutions:
Next-Generation Model Checking and Abstract Interpretation with a Focus on Embedded Control and Systems Biology
Lead PI: Edmund M. Clarke, Carnegie-Mellon University
Collaborators: CUNY, NYU, Stony Brook, University of Maryland, Cornell, Jet Propulsion Laboratory
Computer hardware and software systems can be found in almost all aspects of modern life. While most of us think of computers as the machines on our desks, hardware and software systems embedded in a multiplicity of complex physical systems perform a growing number of important societal functions. For example, embedded computers make our airplanes and cars safer and more efficient, they make our national power grid more reliable, and they provide new diagnostic and therapeutic capabilities in healthcare, ranging from medical imaging systems to implantable heart devices. As these tightly integrated cyber-physical systems perform increasingly complex and important functions, we must engineer them in a way that ensures we can bet our lives on them. This Expedition seeks to explore the use of model checking and abstraction interpretation to analyze and predict the behavior of complex embedded and dynamical cyber-physical systems. The research team will develop the next generation of tools and technologies necessary to enable exhaustive analysis of the behavior of increasingly complex systems, promising safer, more secure embedded cyber-physical systems such as those found in automotive and aerospace applications. Furthermore, the researchers participating in this Expedition intend to use these new tools and technologies to develop transformative systems biology models, promising a deeper understanding of complex biological systems such as inter- and intra-cellular signaling in pancreatic cancer.
Customizable Domain-Specific Computing
Lead PI: Jason Cong, UCLA
Collaborators: Rice, UC Santa Barbara, Ohio State University
Sometimes one-size solutions just don’t meet all our needs. Human civilization has made great advances through specialization, yet most computer users have limited choices when it comes to the type of hardware and software systems they can use to solve a problem in a particular area. The researchers involved in this Expedition believe customized computing has the potential to deliver order-of-magnitude improvements in energy efficiency, development effort, time-to-solution, cost, and overall productivity by crafting computing tools tailored to specific applications and needs. The key to the team’s success is a customizable heterogeneous platform that includes a wide range of customizable computing elements, customizable and scalable high-performance interconnects based on RF-interconnect technologies; highly automated compilation tools and runtime management systems to enable rapid development and deployment of domain-specific computing systems, and a general, reusable methodology for replicating success in different application domains. In the spirit of hardware-software co-design, the research team will balance software and hardware considerations to better expose opportunities for order-of-magnitude improvements in computing efficiency, while using a software approach that supports automation and reuse and is accessible to domain experts. To demonstrate the power of their approach, the team will apply their domain-specific computing design techniques to revolutionize the role of medical imaging and hemodynamic modeling in healthcare, promising cost-effective, yet convenient solutions for preventative, diagnostic, and therapeutic procedures.
RoboBees: A Convergence of Body, Brain and Colony
Lead PI: Robert Wood, Harvard University
Collaborators: Northeastern University
Busy as a bee. A hive of activity. Bees and bee colonies have long been held up as models of efficiency and coordination. Using a host of different sensors, unique communication protocols, and a precise hierarchy of task delegation, thousands of bees can work independently on different tasks while all working toward a common goal–keeping their colony alive. Researchers in this Expedition will create robotic bees that fly autonomously and coordinate activities amongst themselves and the hive, much like real bees. The research team aims to drive research in compact high-energy power sources, ultra-low-power computing, and the design of distributed algorithms for multi-agent systems. Furthermore, the RoboBees created will provide unique insights into how Mother Nature conjures such elegant solutions to solve complex problems.
Basu said that the Expeditions program will continue in the future. “In the last two competitions, the research community has responded with great creativity and enthusiasm. We are now eagerly awaiting the arrival of the next set of Expeditions proposals.”