IEEE: Computing, robotics, education benefit from tighter human-technology interactions
Seven emerging technologies (see photos) could result in automatic guided vehicles that never stop to recharge, more nimble robotics, and smarter pattern recognition for faster cancer detection (and perhaps better predictive controls). Other advances could help those with nerve damage walk using a robotic exoskeleton, then, perhaps under their own power.
Mark T. Hoske, Control Engineering -- Control Engineering, 3/13/2009
|
| Efficient wireless power. WiTricity diagram shows how a power source, left, is connected to ac power. Blue lines represent the magnetic near field induced by the power source. Yellow lines represent the flow of energy from the source to the WiTicity capture coil, which is powering a light bulb. The magnetic field (blue lines) wrap around a conductive obstacle between the power source and the capture device. |
Seven emerging technologies in biomedical engineering, biometrics, computing, robotics, telecommunications, and wireless power, are among world-changing technologies that today's engineers are exploring. For its 125th anniversary, IEEE assembled a panel of experts who shared their work and what it could lead to: automated guided vehicles that never stop to charge, more nimble robotics, and smarter trending for better pattern recognition, which might be applied to advance predictive controls or detect cancer earlier.
One engineering advancement could lead to robotic exoskeletons that help quadriplegics function and, eventually, to signal processors that translate and relay signals between muscles and the brain across a damaged spinal cord or other gap in nerves.
These advances in human-technology interactions, important for human quality of life, can help manufacturing efficiency, too.
Click on each first link below to read more about each speaker’s presentation in the IEEE March 10 webcast, or scroll down and read them all.
After you're done, answer one or all of the TalkBack questions using the tool at the bottom of the page to advance discussion on what these developments mean to you.
- Recharge devices wirelessly: Katie Hall, chief technology officer, WiTricity;
- Predict the future with pattern analysis: Rangachar Kasturi, professor, University of South Florida;
- Math sees cancer as it forms: K.J. Ray Liu, professor, University of Maryland, College Park
- Human-level computing: Dharmendra Modha, manager, cognitive computing, IBM Almaden Research Center;
- Defeat paralysis, walk again: Miguel Nicolelis, co-director, Center for Neuroengineering, Duke University Medical Center;
- Learning slate helps illiteracy: Krishna Palem, professor, George Brown School of Engineering, Rice University; and
- Interoperable, shared electronics: Roy Want, senior principal engineer, Intel Corp.
These and other major efforts aim to inspire engineering. Nicolelis, answering a question afterward, spoke in favor of funding for major scientific projects, for the advancements and benefits they’ll bring, and also to encourage and uplift young people about the excitement of engineering.
“After I saw an American going to the moon, I believed science could do anything,” Nicolelis says. And it seems, science still can do anything, including the following.
|
| Katie Hall, chief technology officer, WiTricity |
Efficient, wireless power
: A more efficient means for wireless power transmission is coming soon, says Hall, from WiTricity. Using magnetic resonance ideas that originated with MIT, milliwatts to kilowatts of power can be transferred wirelessly, recharging devices that are centimeters to several meters away from wired power sources. Possible industrial applications include robotics, mobile human-machine interfaces and control, automatic guide vehicles, roaming diagnostic laptops and handheld instrumentation, and sensors located on rotating devices (avoiding slip rings).
Other applications, Hall suggested, include implanted hearts, electric or hybrid vehicle charging without plugs, and hang-up flat screen televisions without cords.
|
| Rangachar Kasturi, professor, University of South Florida |
Predict the future with pattern analysis
. Kasturi suggests we can help predict the future based on the present and past.
The work can improve prediction of hurricane paths, more accurate code reading, speech, face, medical image, image and video search on the Internet without keywords, and DNA analysis, Kasturi says. Learn more at www.iapr.org and www.icpr2008.org.
|
| K.J. Ray Liu, professor, University of Maryland, College Park |
Better health through math
: Liu created a microarray that translates biological (seemingly random) information in DNA into to an expression of data that can be read. The ensemble dependence model (EDM) looks at proteins broken into small pieces and analyzed, predicting cancer development with 85% to more than 99% accuracy, he says, for cancers of the lungs, stomach, colon, prostate, and ovaries.
|
| A cancer prediction model is identifying when tissues transition to cancer, Liu says. |
The model sees past data that’s normally too noisy to reveal patterns, an application that could have potential in many industrial applications as well, where seeing past noise in data remains important in trend analysis, such as predictive control.
|
| Dharmendra Modha, manager, cognitive computing, IBM Almaden Research Center |
Human-brain capable computers:
Through advances in neuroscience, supercomputing and nanotechnology, researchers are trying to reverse engineer the brain to recreate its power as quickly and economically as possible, Modha says. With real-time simulation of a rat brain cortex possible now, a human brain is only a factor of 400 away, Modha says.
By 2018, human-scale computing will be possible based on projections, but different computing structures would be useful, as biology is much more efficient with space and power.
|
| Modha showed a spreadsheet that represents the cortical functions of a monkey brain. |
Modha showed measurements including neurons, synapses, communication speed, computation power, and memory. While a rat brain operates on 50 mW, a human brain uses about 20 W. The table below provides several other comparisons on computation and memory for a rat, a computer and a human.
| Rat | Computer* | Human | |
| Computation | 78.4TF | 91.8 TF | 38.5PF |
| Memory | 7.2TB | 8 TB | 3.5PB |
| * BlueGene/L, using 32,768 CPUs | |||
| TF is teraflop; TB is terabyte; |
PF is petaflop; PB is petabyte | ||
| Source: Control Engineering |
and | IBM Almaden Research Center via IEEE presentation |
Walk again
: Nicolelis discussed the control loop in humans, as part of the “walk again project,” which eventually could restore movement after paralysis. Sensors in skin and eyes deliver input to the brain, the brain makes a decision, and generates electrical signals that actuate muscles.
|
| Miguel Nicolelis is co-director at the Center for Neuroengineering, Duke University Medical Center. |
He analyzed a monkey brain, and, with orange juice as incentive, taught the monkey to interact with a brain machine interface that let the monkey do tasks on screen by moving a robotic arm with just a thought. Amazingly, the monkey learned to operate the "third arm" with enough brainpower leftover to use its own arms for other things at the same time.
Beyond just having another hand to help out, applications for this technology might first include an exoskeleton that could be operated by paralyzed persons, Nicolelis says. Eventually, signal processing might be sensitive enough for a device to translate brain signals past a break in a nerve to power the human’s own muscles on the other side, in a brain-machine-brain interface, he says. More sophisticated signal processing might feed the brain input from human pressure sensors (touch receptors) in the skin so the brain could again naturally operate a closed control loop, Nicolelis says.
|
| Krishna Palem, professor, George Brown School of Engineering, Rice University |
Literacy efforts
: Palem is using probabilistic electronics to advance literacy and sustainable literacy with what’s called the I-Slate. In India, Palem says, 105,964,434 students have no electricity, and 417,389 primary schools have fewer than 3 teachers. The I-Slate, essentially a solar-powered chalk slate in form factor, is less complex than a laptop and more economical. It uses a probabilistic integrated circuit that analyzes images on a screen from a human’s perceptive and updates the moving image only as the brain requires for clear seeing in the mind’s eye, which, Palem suggests, is more efficient than uniform screen updates.
Interoperable electronics will share resources, if allowed. Suppose the display on a transmitter is trying to tell you something, but you need the big picture? Project the information onto the screen of a nearby portable operator interface. Or, at 2 a.m., when the plant is calling with a problem, extend the browser-based HMI from your laptop onto your flat-screen television, so your bleary eyes can see a little better.
|
| Roy Want, senior principal engineer at Intel, Dynamic Composable Computing |
Roy Want seeks to improve such mobile computing experiences. Why shouldn’t a cell phone, for instance, transmit diagnostics to the appropriate technician on the next continent to resolve the problem and get the manufacturing line running again faster?
IEEE began as the American Institute of Electrical Engineers (AIEE), and Thomas Edison and Alexandar Graham Bell were among founding members. For a lot more about the engineering past and what’s next, see www.ieee125.org.
The webcast was accessed via http://www.ieee125.org/engineering-the-future/media-roundtable.html.
TalkBack questions: Advance discussion on what these developments mean to you by using the tool below to enter your answers.
- Which technologies discussed here could do you the most good in your workplace and why?
- Which development was most unexpected for you and why?
- What other manufacturing, control, automation, or instrumentation applications did you think of for some of these technologies?
Mark T. Hoske, editor in chief
Control Engineering News Desk
Register here and scroll down to select your choice of eNewsletters free.
Wireless power, power harvesting
02/09/2010International AC power source
02/09/2010Power-over-Ethernet switch
06/01/2005
View All Webcasts
