A world-changing view of engineering

A world-changing view of engineering

Energy, transportation, environment, health care, third-world poverty-the list may sound more like the agenda of an international relief agency than an engineering department. But the Center for Information Technology Research in the Interest of Society is attempting to tackle just such problems with a multidisciplinary engineering approach, one that combines information, biological and nano technologies.

Based at the University of California at Berkeley, the center, known as Citris, is coming into its fourth year with some 150 research projects, more than 200 participating faculty at four UC campuses, and several hundred million dollars in state, federal and corporate funding. Its charter is to tackle society's toughest problems with leading-edge, IT-based solutions.

Citris is perhaps best-known for its development of "smart dust" motes, tiny wireless sensors that have been installed in buildings for energy efficiency, in bridges for seismic safety and in redwood trees to monitor environmental conditions. But the scope of Citris projects is incredibly diverse, ranging from a global data storage system to fingernail-size syringes, interactive historical maps, oceanic monitoring, micromechanical flying "insects" for search and rescue, and a scalable IT architecture for the third world.

All are examples of what Citris calls societal-scale information systems, which aim to solve large-scale problems. Such a system integrates devices ranging from tiny sensors and actuators to handheld information appliances, workstations and computers. They're connected by short-range wireless networks as well as high-bandwidth, long-haul optical backplanes.

In a broader sense, Citris represents a radical new approach to engineering, said Richard Newton, dean of engineering at UC Berkeley and a key Citris founder. "Engineering is about applications-it's about fundamental research inspired by a use," Newton said. Rather than take a compartmentalized approach to electrical, mechanical or civil engineering, he said, the project challenges students and faculty to identify a problem and then acquire the skills to solve it.

Newton called Citris "an experiment in a new way of doing research." He noted that yesterday's big research institutes, like Xerox Parc, have either disappeared or lost prominence. He views Citris, with its backing from a number of large corporate donors, as a likely heir to those institutes.

Indeed, Citris appears to be giving birth to some leading-edge research. Earlier this month, one Citris project came to fruition as UC Berkeley and Stanford University researchers announced the creation of the first working, integrated silicon circuit that incorporates carbon nanotubes into its design.

While most of the Berkeley faculty members involved in Citris hail from electrical engineering, the project also encompasses faculty from mechanical and civil engineering, as well as social sciences, humanities and law. "We are changing how you bridge different disciplines, and not only engineering," said Ruzena Bajcsy, Citris' director. "This is about problem solving. This whole experiment is about changing the culture."

"I think what we're really trying to do with Citris is to have the IT revolution not be a disruptive element in societal change but a constructive element," said Shankar Sastry, who chairs Berkeley's department of electrical engineering and computer science.

According to Newton, the ideas behind Citris began to jell in 1999, when faculty at Berkeley's College of Engineering began to talk about the future of engineering. They came to feel that a multidisciplinary approach was essential and that the existing departmental structure was not conducive to such an approach.

"We want to say to a student, instead of choosing to be an EE or a mechanical engineer, choose a problem, like environmental cleanup," Newton said. "We still need departments, and we will never compromise on fundamentals, but our long-term goal is to make the core college interdisciplinary."

In late 2000, former California Gov. Gray Davis decided to fund three science and technology institutes that would leverage research from the University of California. Three were chosen, but the Citris proposal, which included faculty from the Berkeley, Santa Cruz, Davis and Merced campuses, so intrigued the governor that Citris became the fourth funded institution.

The award brought in $100 million in state funds, most of which will be used to construct a Citris headquarters at UC Berkeley that will include a nanofabrication facility. Citris also received pledges for around $250 million from private sources, industry, and federal grants and contracts. Some of that money has already been received and spent. Everything Citris develops is placed in the public domain.

Founding corporate members include BroadVision, Ericsson, HP, IBM, Infineon, Intel, Marvell, Microsoft, Nortel, STMicroelectronics and Sun Microsystems. "Citris is an attempt to define a new model for academic and industry cooperation," said Patrick Scaglia, director of the Internet and Computing Platform Research Center at Hewlett-Packard Co., which is collaborating on a number of Citris projects. "All the work is open to everyone, and the IP [intellectual property] is no longer encumbered."

Mighty motes
Funding and mission in hand, Citris researchers went to work attacking problems in six key areas: energy efficiency, transportation, health care, education, emergency response, and land and environment. Since then, other areas of focus have been added, such as service to developing countries.

Work on smart-dust motes had already begun at UC Berkeley under a Defense Advanced Research Projects Agency contract. Among the first Citris developments was Mica (see story, below), a half-dollar-size mote based on two pc boards and on TinyOS, a comms-centric operating system designed for very small memory footprints.

Mica and TinyOS are available on an open-source basis, and variations of both are in use in academia and industry. A private company, Crossbow Technology, has developed commercial versions of Mica. The next step is chip-level motes, and back in the lab, Citris researchers have developed wireless-sensor chips as small as 5 square millimeters. The long-range vision is one of microscopic, self-powered motes.

Within Citris, distributed networks of Mica-based motes have found numerous applications. They have been used to make buildings at UC Berkeley more energy-efficient, to measure vibration on the Golden Gate Bridge, to monitor a petrel seabird colony in Maine, to track the growth of wildfires and to measure ground liquification in Japan. One Citris project is using wireless sensors to develop smart helmets for Chicago firefighters.

Networks consisting of millions of tiny, low-cost wireless sensors could solve many problems in developing countries. For example, Newton noted, smart-dust motes with biological sensors could be dropped into ponds and rivers to detect malaria bacteria and send a signal upondetection, so that spraying could commence at exactly the right time.

"This is a problem that unites bio, info and nano in one application, inspired by a use," said Newton. "It motivates students like never before."

There are many other broad-ranging Citris projects. One seeks to develop a highly redundant, global data storage system so that data can never be lost or compromised. Another is designing high-frequency radar technology to track ocean currents and wind velocity. Yet another is working on solar-powered, insectlike flying robots that can be used for emergency response, monitoring and reconnaissance.

Citris, meanwhile, has given birth to the ICT4B initiative, which is attempting to find a scalable information and computing technology architecture for the 4 billion people globally who make less than $2,000 a year.

Part of this effort involves the design of low-cost, handheld computing devices with user interfaces targeted at the illiterate, said Vivek Subramanian, professor of electrical engineering and computer science at UC Berkeley. They could, perhaps, tell a farmer what his crops are worth, or network with sensors that could detect arsenic in well water. Subramanian said Citris plans an initial deployment in India.

"I totally disagree with people who say the IT revolution is over," Newton said. "Now is the era when we're going to apply these technologies to transform the world in dramatic ways."

Citris online
- Citris overview: www.citris-uc.org

- TinyOS: http://webs.cs.berkeley.edu/tos

- Crossbow Technology: www.xbow.com

- Dust Inc.: www.dust-inc.com

http://www.eet.com