December 2005
Jose Carmena is also affiliated with UC Berkeley and UC San Francisco's Joint Graduate Group in Bioengineering. |
For decades, the ability to control computers and robots with our minds has been the stuff of science fiction. Recently though, UC Berkeley professor Jose Carmena has made strides to bring the underlying technology into the real world. His research on brain machine interfaces could someday enable disabled individuals to be fitted with bionic prostheses operated by thought alone.
"Many people with severe physical disabilities have intact brains," says Carmena, a professor of electrical engineering and computer sciences, who joined the UC Berkeley faculty this year. "If we could extract and decode the intention of the patient from their brains, the signals could control any device, from a robotic arm to the cursor on a computer screen."
Recently, Carmena was part of a team at Duke University that famously demonstrated how rhesus monkeys could learn to operate a robot arm with brain signals. The commands came via an array of electrodes implanted in the frontal and parietal lobes of the animal's brain. In 2003, the researchers made headlines by showing that the monkey wasn't treating the robot as an external device. Instead, its brain's structure had adapted to control the appendage as if it were its own arm.
The implant consists of an array of several hundred hair-thin electrodes surgically placed in the frontal and parietal lobes, regions of the brain involved in motor abilities. Each microwire can detect the signals from as many as four neurons. Last year, Carmena and his colleagues recorded electrical signals from a human patient's brain. The array was temporarily implanted during deep-brain stimulation surgical procedures conducted to alleviate tremors symptomatic of Parkinson's and other diseases.
"The main benefit of using an array of many microelectrodes is that you can observe the patterns of how signals move through the brain over time," says Carmena, who is also affiliated with UC Berkeley's Group Major in Cognitive Science and the Helen Wills Neuroscience Institute. "From a neuroscience perspective, visualizing these neural functions helps us study how the brain learns and adapts."
An array of hair-thin electrodes that the researchers use to record electrical signals from the brain. [View larger image] |
Of course, the brain-machine interface depends on nothing getting lost in the translation between thought and robotic action. At Berkeley, Carmena is attacking all the layers in the interface between mind and machine. Firstly, he will collaborate with researchers from the university's state-of-the-art microfabrication facilities to develop better electrode arrays. Currently, s mall movements of the recording electrodes in the cortex often damage the neurons, weakening the signal over time.
"So far, we've recorded signals for two years, but in a human patient you'd want the electrodes to work for much longer," Carmena says. "We need a device as dependable as a pacemaker, but for the brain."
Along with new electrodes, Carmena and his colleagues are improving the brain-machine interface itself. The aim is a small device that requires very little power but can handle the large bandwidth of brain signals. Then, once the signals are routed to the computer, advanced software algorithms are required to distinguish the signal from the noise. Finally, they must design and test novel prosthetic devices, from robot grippers and exoskeletons to software that could enable patients with neurological disorders such as locked-in syndrome to interact with the external world via computer.
Eventually, Carmena says he'd like to close the loop on brain-machine interfaces beyond just observing what's being controlled. Feedback, he explains, is essential for proprioception, knowledge of our body's place in the physical world. For example, proprioception is what enables us to touch a finger to our nose even with our eyes closed.
"We'd like patients to feel where the artificial arm is in space and perhaps even experience the sense of touch," Carmena says. "We've shown that we can extract signals from the brain, but can we encode information back into it?"
Nature's Nanoshells
by David Pescovitz
William Holtz is a graduate student in the Department of Electrical Engineering and Computer Sciences. |
Taking inspiration from nature's elegant engineering, UC Berkeley graduate students are working to create novel nanoscale structures modeled after a common marine organism. Using techniques pioneered at the Berkeley Center for Synthetic Biology, the students hope to produce designer materials resembling in form and function the tiny intricate shells of photosynthetic algae called diatoms. Initially, the biomimetic diatoms could be employed as filtration systems or self-contained catalysts for a lab-on-a-chip used for medical testing. Eventually, the structures could enable the fabrication of more powerful computer chips containing circuits patterned in three dimensions or act as substrates for the in vitro growth of human tissue for implantation.
"As an electrical engineer, I come at this work from a microfabrication perspective," says William J. Holtz, a Ph.D. candidate in the Department of Electrical Engineering and Computer Sciences (EECS) who collaborates on the work with Afshan Shaikh, a Ph.D. candidate in the Department of Chemical Engineering, EECS Ph.D. candidate Frank J. Zendejas, and others . "The question we're asking is how can we create very small systems that self-assemble into desired structures?"
Afshan Sheikh is a graduate student in the Department of Chemical Engineering. |
Holtz and Shaikh are advised by Jay Keasling, professor in the Departments of Chemical Engineering and Bioengineering and director of Berkeley Center for Synthetic Biology. Keasling specializes in genetic techniques to convert bacteria into chemical factories that produce, for example, the precursors of anti-malaria and anti-cancer drugs.
The allure of diatoms is that the shells contain detailed features that are just a few nanometers in size. (A nanometer is one billionth of a meter.) Today, only the most experimental lithography tools are capable of patterning features of comparable scale on silicon, and even then only in two dimensions. Furthermore, those methods aren't suitable for large-scale industrial production.
"Diatoms are living organisms though, so they produce the structures at ambient temperature and pressure at very high rates," says Holtz.
Scanning-electron micrograph of a species of diatom. [view larger image] |
If the researchers can devise a method to mimic the natural generation of the shells, they could potentially make the materials shaped-to-order. For example, a shell with very specific pore sizes would enable some molecules to flow through while keeping others out, acting as a nanoscale filter. Adding other chemicals to the interior surface of the shell could transform it into a miniature test tube containing a built-in catalyst. These kinds of structures would be useful as the basis of, say, an ultrasensitive detector of specific gases or a handheld device for molecular medical diagnostics in the field
" These kinds of devices require a lot of chemistry," Holtz explains. "Diatom-like structures would be very helpful in that."
The synthetic diatoms might also be used as templates for gears smaller than a period, as miniscule lenses, or to bulk manufacture 3D circuits and silicon micromachines that today must be painstakingly fabricated layer by layer. The key is to be able to crank out the synthetic shells in bulk in specific shapes. Holtz, Shaikh, and their colleagues are exploring several possible methods.
Scanning-electron micrograph of another species of diatom. |
Holtz is attempting to extract diatom proteins responsible for the shell production and plans to use off-the-shelf polymers as the raw materials in the self-assembly process. Tiny amounts of the ingredients mixed in a solution chemistry process could then produce the shell-like structures. But even if the technique is successful, Holtz says, the protein extraction isn't practical for industrial applications.
To that end, Shaikh is attempting to genetically re-engineer bacteria into a microbial factory that cranks out the proteins cheaply and easily. She's working on transferring all of the genes that encode the shell-making protein's metabolic pathway into E.coli. The next step is to get the gene to express itself so the E.coli can actually start producing the protein.
"You could then take those proteins and do the in vitro process to make the shells," Shaikh says.
According to the researchers, the structures are well-suited substrates for the in vitro growth of cells and tissues that could later be implanted into humans to heal injuries. Indeed, someday the synthetic shells may be found inside everything from our bodies to next-generation PCs to handheld medical laboratories.
"Even more intriguing is the possibility of self-repairing systems, in which these low-energy biosynthetic techniques would enable the re-growth of damaged devices in the field," the researchers say.
Ken Goldberg is affiliated with CITRIS and the editor of two books on Internet telerobotics, both published by MIT Press: The Robot in the Garden and (with Roland Siegwart) Beyond Webcams. (Bart Nagel photo) |
Almost every night, cars are broken into in Yosemite National Park. It's classic smash-and-dash. Windows are shattered and goods stolen. It's not a human crime ring though. The thieves are black bears and they're mostly after food that visitors leave in their vehicles. Soon though, a new telerobotic surveillance system that enables visitors to "tour" the park via the Internet may also help capture footage of the bear burglars. The installation would be a proof-of-concept test for the Collaborative Observatories for Natural Environments (CONE) technology that UC Berkeley robotics professor Ken Goldberg is developing to aid scientists studying natural animal behavior in remote places.
"Biologists spend a great deal of time observing and recording nature using traditional video equipment," says Goldberg, who holds a joint appointment in the Department of Industrial Engineering and Operations Research and also Electrical Engineering and Computer Sciences. "So we're trying to help them bring the latest technology into the field."
A prototype CONE in Goldberg's laboratory. |
Right now, the study of animals in the wild over long periods can be difficult, expensive, and sometimes dangerous. For example, scientists would like to watch families of Alaskan polar bears emerging from their dens in the spring. While it's fine weather for bears, humans aren't accustomed to the whipping winds and freezing cold.
Working with his former graduate student Dezhen Song, now a professor at Texas A&M University, Goldberg is designing robotic "observatories" that scientists could leave behind at their research sites. Once they return to the laboratory, they could log on to the Internet to see what the camera see and steer it to keep an eye on their animal subjects from afar. The system draws from advances in high-resolution robotic cameras, long-range wireless networking, distributed sensor networks, and software algorithms for collaborative control that Goldberg and Song developed in the last few years.
The idea, Goldberg explains, is that the CONE would be contained in a small, wheeled trunk. After opening the lid, the system automatically kicks into operation, seeking out a satellite connection for Internet access and charging its batteries via solar panels. Meanwhile, the scientist distributes a handful of small, wireless sensors (pioneered at UC Berkeley) that monitor motion, temperature, and other variables. T he sensors self-organize into an ad hoc wireless network and pass their data from one to another, bucket-brigade style until the information reaches the CONE.
Depending on what the sensors detect, the telerobotic camera can then point itself at the source of the activity – for example, unusual movement off to one side of the panorama. The system might then create a time-lapse clip of that specific region for later review by the scientist.
What happens if more than one sensor calls for the camera's attention, or more than one scientist commands it to move simultaneously? That's when Goldberg and Song's advanced control algorithms kick in. The software uses mathematical principles to infer a consensus from a group of requests. That way, the observatory can point the camera in the direction that will satisfy the most users all of the time. They're also working on a time-based technique for the camera to respond in turn to each request as efficiently as possible.
"It's a hybrid system though, so it's collaborative not just among people but also among sensors," Goldberg says. "A change in the image can be requested by people or sensors and that request can be weighted depending on who or what is making it. If the chief biologist wants to see something, her request can override everything else."
Currently, Goldberg and Song are working with the National Geographic Society on a plan to test their prototype CONE in Yosemite in the near future. They've also begun to discuss more scientific collaborations with UC Berkeley biologists. In the short term though, the technology may soon save visitors from facing the reality that animals can be criminals too.
"The rangers want to show people what can happen if they leave food in their cars," Goldberg says. "Of course, viewers safe at home may enjoy watching the brutes at work: Go Bears!"
Jonathan Cagan (Ph.D.'90 ME) |
You're an engineer on a product design team. Your company is under intense pressure from foreign competition. Management asks you to come up with a Hail Mary product, something to push the company ahead. You must innovate. How to start?
You could rub a lucky rabbit's foot and wait for ideas to come. But you're an engineer. You want a plan for this elusive thing called innovation. A good place to start is ME alum Jonathan Cagan's new book, The Design of Things to Come: How Ordinary People Create Extraordinary Products. Cagan and co-authors Craig Vogel and Peter Boatwright give tools, strategies, and a methodology for early product design. The three authors offer plenty of real-world examples showing how engineers and designers used these processes to come up with innovative products.
"A large number of new products fail" says Cagan (Ph.D.'90 ME). "We can't guarantee success, but we've shown that if you follow a good process, you increase your chances of success. Innovation is the result of hard work."
Cagan is an ME professor at Carnegie Mellon, focusing on early design processes. Much of the content for the book came from his teaching, research, consulting, and from starting his own company. Published in June, the book has already received warm praise. "It disaggregates the broad concept of 'innovation' into usable ideas and strategies that can be implemented" writes Bruce Nussbaum, editorial page editor at Business Week. "I learned a great deal about innovation and design from it."
Writing the book wasn't easy, Cagan says. Each author drafted several chapters and shared them with the others for revisions. "There were a lot of arguments and late nights, but we tried to be an example of what we talked about in the book in terms of real integration between disciplines." Cagan's co-authors include a marketing professor and a design professor.
As a Ph.D. student studying under ME professor Alice Agogino, Cagan worked long hours on computational design but budgeted plenty of time for other things. He was president of Cal's ballroom dance club and enjoyed Berkeley's restaurants and coffee shops, SFMOMA, San Fran-cisco, and the beaches. "It's a really invigorating area to live in" he says.
With increasing pressure on engineers to deliver the goods or turn in company key cards, Cagan has this advice for students: "You need to know engineering fundamentals but recognize that technology alone isn't the answer. Take classes in art history, architecture, or business and read Fast Company or Business Week. All this will help you think in a broader context about how technology fits into rest of world."
Vogel, C. M., J. Cagan and P. B. H. Boatwright, The Design of Things to Come: How Ordinary People Create Extraordinary Products, Wharton School Press/ Prentice Hall, Upper Saddle River, NJ, 2005.
