E-textiles:
Creating the future’s wearable, washable, potentially life-saving computers
By Liz Crumbley , College of Engineering
When Tom Martin came from the University of Alabama in Huntsville to Virginia Tech in 2001 to interview for a position as an assistant professor in the Bradley Department of Electrical and Computer Engineering (ECE), he found a novel research calling as well as a new academic home.
Martin met Mark Jones, an associate professor of ECE who had secured funding from the Defense Advanced Research Projects Agency (DARPA) for a project called STRETCH. Jones was working with researchers at the University of Southern California’s Information Sciences Institute in Arlington, Va., to develop large electronic textiles (e-textiles) fabrics that would look like typical tents or camouflage nets, but that could detect the faint sounds of distant vehicles being deployed by the enemy.
As Jones and Martin talked about STRETCH, they began to realize that e-textiles could be a giant step forward in designing wearable computers. At that time, a typical wearable computer was a small CPU in a fanny pack connected to cumbersome headgear that held a display screen at eye level. Jones and Martin envisioned something entirely different.
Because the wires and sensors in e-textiles are woven into the fabric, wearable computers made of e-textiles can be constructed much like normal-looking shirts or hats or other types of cloth apparel. These computers do not have large screens and bulky keyboards, but would perform specific functions necessary to the wearers.
“Wearable computers constructed of e-textiles can offer context awareness,” Martin says. “They can be designed to be aware of the user’s motions and surroundings.”
Infant technology, giant steps at Virginia Tech
The brief history of e-textiles dates back to the late 1990s, when a group of students at the Massachusetts Institute of Technology (MIT) created some artistic applications for electronics, such as balls that play music and drapes that change colors when their sensors are alerted.
A few other universities, including Georgia Tech and the University of California at Los Angeles, also have e-textiles research programs, and a handful of companies are working on electronic clothes and sports equipment. Textronics Inc., for example, manufactures a sports bra outfitted with electrodes that can monitor heart rates during athletic training. Adidas makes sneakers with tiny computers in the soles that can adjust cushioning as the wearer runs.
But if you launch a Google search on “e-textiles,” you’ll likely find more hits for the Virginia Tech E-textiles Laboratory, founded by Jones and Martin, than for any other single entry.
When Martin joined the ECE faculty in fall 2001, he and Jones began a successful collaboration in e-textiles for wearable computing. Within a year, while working together on the STRETCH project, they won a National Science Foundation (NSF) Information Technology Research (ITR) grant supporting their development of hardware and software architectures for the design of “tailor made” e-textiles fabrics.
If tiny electronic monitoring devices could be woven into everyday attire, then cardiac patients wearing e-textiles vests could comfortably have their blood pressure and heart rates monitored as they went about daily routines. Athletes could wear pants that would collect data about their speed and running gait characteristics. Firefighters could enter burning buildings in uniforms that would collect information and map directions for safe return routes.
Weaving the Hokie Suit
The Virginia Tech colleagues bought a hand loom. Another faculty member’s wife, Dana Reynolds, who is experienced in operating a loom, helped create a “Hokie Suit” for the project, weaving thin, flexible wires among orange and maroon cloth threads to form a vest and pants that could be outfitted with electronic sensors.
Josh Edmison, a graduate student in the e-textiles lab, helped design the Hokie suit and researched the technical aspects and potential uses for his master’s thesis.
Meanwhile, Thurmon Lockhart of Virginia Tech’s Grado Department of Industrial and Systems Engineering (ISE) was busy in his Locomotion Research Laboratory, trying to unlock the biomechanical mysteries of human slips and falls.
With funding from the Centers for Disease Control and Prevention and the National Institutes of Health, Lockhart and his students had devised an experimental setup — a walking platform, harness apparatus, and system of sensors — that enabled them to monitor muscle and joint activities in the feet, ankles, legs, hips, and arms. Data fed into a computer model provided the researchers with data about each volunteer subject’s walking gait and body motions during slips and recoveries.
When they learned about Lockhart’s project, with its primary goal of determining the causes of slips and falls among older people, the e-textiles group realized that their Hokie Suit could be equipped for gait analysis experiments in the locomotion lab. Edmison, Jones, and Martin worked with Lockhart to set up the Hokie Suit with a “gait matrix” of various sensors — such as accelerometers, which detect changes in speed and direction of motion. Edmison and other students donned the suit and harness in Lockhart’s lab and began collecting data.
“Our goal is an e-textiles suit outfitted to perform gait analysis that can detect the types of changes in gait that would signal potential falls,” Martin says.
Someone prone to falling could wear pants woven with a network of wires connecting sensors, which could communicate with each other and with a data-collecting computer. The wearer could walk around naturally while gait measurements and other data were being collected. The ultimate goal would be to devise a system that would gather data and then alert the wearer instantaneously if he were in danger of falling.
A multitude of medical uses
Another example of a beneficial use for this combination of e-textiles and locomotion research, Martin says, would be testing the movements of children suffering from cerebral palsy. “Doctors could collect useful data while children were at home and moving naturally, rather than in a hospital lab.”
E-textiles clothing like the Hokie Suit would have a number of research and medical applications aside from gait analysis. For instance, the Virginia Tech researchers are perfecting methods of constantly monitoring the activities of cardiac patients while measuring their heart rate, blood pressure, and temperature. “Knowing what a patient is doing could help a doctor interpret heart-rate data,” Martin says.
Other medical applications could include detecting when a wearer has fallen down and can’t get up or has had a seizure – and then sounding an alarm for help.
Environmental sensing is another use Martin imagines as part of the future of the technology. “E-textiles clothing could be worn by people who suffer from asthma, for example. The electronics could be set up to figure out what in a certain environment causes – or might cause – an asthma attack,” he says.
Wireless transmitters placed on garments could communicate data to a screen, a cell phone, or a computer – or via cell phone to a doctor’s office or emergency alert system.
E-textiles garments could also be used by athletes to collect a variety of data for feedback during training and sports competitions. Potential industrial uses include collecting data on the motions of workers in manufacturing plants or at construction sites to help determine why certain injuries are occurring.
Perhaps one of the most popular everyday uses would be e-textiles garments for infants — to monitor sounds, heart rate, motion, location, and other conditions of concern to parents.
As a prototype for medical monitoring, the Hokie Suit was designed for easy use. The wearer can move around naturally and the wiring and electronics can be woven and fitted so that they seem part of the fabric. The electronics are powered by a regular nine-volt battery carried in the pocket. And, as the ultimate note of practicality, the Hokie Suit is washable.
“You can quickly remove the sensors and then wash the suit with the wiring left in,” Martin says. “In a production version, most of the electronics could be sealed so they wouldn’t have to be removed when the suit is washed.”
In hopes of continuing and expanding their e-textiles locomotion study, Jones, Martin, and Lockhart submitted a proposal to NIH. To carry out research on the scale they proposed, they knew several e-textiles garments would have to be woven, many more than could be produced on the hand loom. Through an NSF Computing Research Infrastructure (CRI) grant they had received, Jones and Martin purchased an automated and computer-controlled industrial loom.
In addition to the NSF ITR and CRI grants, the researchers have received funding from Intel. In 2005 Martin won a Faculty Early Career Development Program (CAREER) Award grant, which was followed the next year by a Presidential Early Career Award for Scientists and Engineers (PECASE).
Industrial loom and simulation software
Martin and Jones — who in the mid-1990s won a CAREER award for work in configurable computing while he was an assistant professor at the University of Tennessee — also wanted to investigate the potential for mass-producing e-textiles for medical, fitness, and safety monitoring applications. Although the researchers were experts in the field of computing, they were rank amateurs in matters of loom operation and textiles weaving. But once again, they were in luck when it came to recruiting graduate students.
Meghan Quirk had begun her undergraduate studies at Philadelphia University, where for two years she majored in textiles engineering. She then transferred to the University of Maryland to work on a bachelor’s in computer science. Before graduating, she visited Virginia Tech as part of a Research Experience for Undergraduates. Having read about some of the work Jones and Martin were doing, Quirk met with them and they invited her to join their research group as a graduate student after she finished at Maryland.
Quirk did know about looms and textiles weaving. She chose a suitable industrial loom and, after the manufacturer set it up in the e-textiles lab in Torgersen Hall, she programmed the machine for weaving.
With the funding from Intel, Quirk began weaving an e-textiles rug. “The goal is to make a simple demo rug that will change light patterns as people walk on it,” she says. Into the rug’s fabric Quirk has woven stainless steel and piezoelectric wires that can sense where a person is standing, electroluminescent wires that produce light, and insulated metal wires that connect all of the electronic elements.
“We’re weaving the rug and e-textiles garments as prototypes to demonstrate the elements and principles of the technology,” Martin says. Their central research focus is to develop computer hardware architecture and software environments for creating successful e-textiles applications. If they succeed, designers will be able to easily create e-textiles garments, rugs, and other items for medical, military, and industrial uses.
However, the central and most difficult aspect of the research is figuring out how to develop simulation environment software for e-textiles production. The simulation software will mimic the way in which various electronic elements should be placed for optimal performance.
Quirk’s thesis is focused on discovering how to weave fabric so electronic elements are in the right places along the network of wires. “Weaving e-textiles fabric is a small part of showing that our ideas for simulation and software will work,” she says.
The trickiest problem they must solve on their way to creating simulation environment software is figuring out how to arrange networks of electronic elements so that a garment “can keep track of its own shape and know what the human inside is doing,” Martin says.
To give a simple example: if an e-textiles shirt is supposed to monitor the movement of a wrist and the wearer pushes up the sleeves, the electronic elements need to be aware that monitoring should be temporarily suspended — or they need to re-route their surveillance to another part of the garment so that monitoring can proceed uninterrupted.
Perfecting a hardware and software system to handle such complexities won’t be easy. For one thing, the researchers are dealing with an array of sensors — including accelerometers, gyroscopes that measure rotational velocity, microphones, ultrasonic emitters and detectors that pinpoint location, and piezoelectric films that change voltage in response to shape change.
The benefits that e-textiles have the potential to provide in medical and safety applications will depend on the successes of the Virginia Tech group and some like-minded researchers in the U.S., Europe, and South Korea.
“All of the applications we’ve thought of for this technology are possible,” says Martin. “That is, assuming we can create e-textiles fabrics that can detect their own shape.”
