Ten years ago, Evans suffered a stroke that left him with limited mobility. Over the past two years, he’s been working with Kemp, an associate professor in the Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory University, to develop and test robots that help him shave, adjust a blanket when he’s cold, and even scratch an annoying itch.
“We did things with the robots that I never could have imagined,” said Evans, who contacted Kemp after seeing him on a CNN broadcast about health care robots.
Robots working directly with people – even helping them shave – is both challenging and unusual. Most robots today work in manufacturing facilities where, for safety reasons, they stay far away from humans. But Georgia Tech robotics researchers believe people and robots can accomplish much more by working together – as long as the robots have common sense to know, for instance, how much force humans apply when shaving.
“A major challenge for health care robots is that they lack so much of the knowledge and experience that people take for granted,” said Kemp. “To us, it’s just common sense that everybody has; for robots, it’s a serious impediment.”
Giving robots common sense is just one milestone on the path to the kinds of collaboration that will be required to meet the needs of a growing population of older persons. Beyond personal care, the benefits of co-robotics are many. To produce better products more efficiently, manufacturing robots will need to team up with humans, each contributing unique abilities. And in defense and homeland security, robots will increasingly have to take on the dangerous jobs, leveraging people’s skills while protecting them from harm.
Read more of this article from Georgia Tech's Research Horizons magazine.
]]>Research News
(404) 894-6986
]]>“Our faculty leverage robotics research across many disciplines,” said Steve Cross, Executive Vice President for Research at Georgia Tech. “Their combined resources and expertise are leading to practical solutions for many critical challenges in areas of science and engineering.”
The projects received funding through the National Robotics Initiative (NRI) program, first announced by President Obama in June 2011, and led by NSF with support from NASA, the National Institutes of Health, and the USDA. The purpose of this program is to develop the next generation of robotics, to advance the capability and usability of such systems and artifacts, and to encourage existing and new communities to focus on innovative application areas.
Tech researchers received more than $2.8 million in NRI funding in 2012 and 2013, bringing the combined total of all current awards to over $5.75 million.
“Georgia Tech has a consistent record of achievement in the field of robotics,” said Henrik I. Christensen, KUKA Chair of Robotics and executive director of the Institute for Robotics and Intelligent Machines (IRIM), the flagship for Tech’s robotics efforts. “I’m proud of our amazing faculty, our excellent research, and the awards our faculty members have received.”
Four projects received funding through the Information & Intelligent Systems (IIS) division of NSF’s Directorate for Computer and Information Science and Engineering:
The fifth project, “Multipurpose Robotic Platform for Field Scouting and Sampling,” received more than $900K from the USDA. Led by IRIM Associate Director of Research and GTRI Principal Research Engineer Gary V. McMurray, the research aims to improve crop yields by developing the robotic technology required to autonomously collect data to monitor crops in a field or orchard, including crop scouting for biotic yield-reducing factors, such as pest organisms like insects, plant pathogens, and weeds; as well as abiotic stresses, such as inadequate moisture and nutrient levels. Professor Glen C. Rains (Department of Entomology at the University of Georgia) serves as co-PI on the project.
Robotics research at Georgia Tech attracts more than $35 million in sponsored research each year. Core research areas include mechanisms, control, perception, artificial intelligence, human-robot interaction, and application systems technologies. Tech advances personal and everyday robotics through its research into the ways robots can learn from and interact with humans, by developing advanced technologies, and by exploring issues surrounding the governance and ethical use of robots.
This research is supported by the National Science Foundation (NSF) under Awards IIS-1426998, IIS-1426443, IIS-1426945, and IIS-1427300; and by the United States Department of Agriculture (USDA) under Award 2014-67021-22556. Any conclusions or opinions are those of the authors and do not necessarily represent the official views of the NSF or USDA.
]]>College students from across the U.S. are getting an opportunity that is unavailable to many of them on their own campuses—the chance to program robots. The students are taking part in a 10-week Summer Undergraduate Research Experience, or SURE for short.
This particular SURE program is at Georgia Tech’s Institute for Robotics and Intelligent Machines (IRIM). It’s an immersive robotics research experience that is designed to attract underrepresented students into graduate school in the fields of engineering and science. IRIM associate director of research Ayanna Howard says the projects—such as programming a robot to serve food to a person with a disability—are also meant to enlighten the students about the various applications of robotics and the multidisciplinary aspects of the research. She says the students also develop more confidence.
In addition to the hands-on technology, the students also get advice on how to apply to and pay for graduate school. SURE is supported by the National Science Foundation’s (NSF) Research Experiences for Undergraduates (REU) program, and co-funded by the Department of Defense (DoD) in partnership with the NSF REU program and NSF's Division of Engineering Education and Centers (EEC).
The research in this episode was supported by NSF award #1263049, REU Site: Summer Undergraduate Research in Engineering (SURE).
]]>Georgia could become the fifth state to make self-driving cars street legal.
By opening up its roads and highways to autonomous vehicle testing, Georgia would tap jobs and investment from an ecosystem of tech firms and automakers developing self-driving technologies.
Legislation that could come before the General Assembly this winter would provide opportunities for Atlanta companies that develop auto-related software and services. Fleets of self-driving cars — which wirelessly communicate with each other to manage traffic flow — could be the answer to Atlanta’s notorious gridlock.
“Over time, this is the real game changer we’re looking for,” said state Rep. Ed Setzler, R-Acworth, a member of a House of Representatives study committee on autonomous vehicle technology that will begin meeting in a few weeks.
The committee will establish a legal structure for assessing fault in case of a crash involving a self-driving car, Setzler said. “If a car is on the street and a driver is not behind the wheel, a chain of responsibility needs to be in place so if something goes badly, things will be made right,” he said. The committee will make recommendations to the General Assembly by year-end.
Some luxury cars already boast semi-autonomous features, including automatic parallel parking, lane-departure alerts and adaptive cruise control.
“The self-driving car is the ultimate response to distracted driving,” said Thilo Koslowski, Gartner vice president. By 2016, at least three companies will announce launch dates for their first self-driving car models, he said. Carmakers are interested in reducing traffic accidents and keeping the automobile relevant. “When we have an aging population and younger consumers that aren’t interested in owning a vehicle, the auto industry needs to figure out different ways to attract customers,” Koslowski said.
Traffic congestion, auto accidents and new business opportunities are driving momentum for the development of autonomous vehicle technologies. A third of U.S. vehicle owners surveyed by Gartner in the first quarter said they are interested in having self-driving capabilities in their next vehicle.
Georgia, a transportation hub and research center, is suited for development of self-driving technologies, said Henrik Christensen, executive director of Georgia Tech’s Institute for Robotics and Intelligent Machines.
“Our infrastructure is already extended beyond capacity and [autonomous driving technologies] offer a great opportunity to address that,” he said.
Georgia Tech is doing research on autonomous driving technology, smart sensors, car safety systems and increased driver awareness.
“Large-scale experiments to explore interaction between regular cars and autonomous cars [are] interesting,” Christensen said. “As an urban campus, Georgia Tech is ideally suited to explore this.”
Michigan, California, Nevada and Florida have taken the lead — opening up to self-driving vehicle research.
Self-driving vehicles will be transformative for commuting, noted Stephen Fleming, a Georgia Tech vice president.
Autonomous vehicle development is being steered by the tech industry.
A lot of the processing is not going to happen in the car, but in the “cloud” and between vehicle systems.“It’s not ‘do we know how to build an engine and a transmission,’” Fleming said. “It’s ‘do we know how to build software networks that can negotiate with other networks in a secure and rapid manner’ — and that Atlanta knows how to do.”
]]>The A. Richard Newton Educator ABIE Award recognizes the development of innovative teaching practices and approaches that attract girls and women to computing, engineering, and math. A robotics researcher, teacher, and mentor, Howard has engaged hundreds of female and minority students in computing, engineering, and science through numerous K-12 outreach programs and summer camps.
Before coming to Georgia Tech, Howard worked at NASA’s Jet Propulsion Laboratory (JPL) at Caltech from 1993-2005, where she led research efforts on various robotic projects that led to her being named to the MIT Technology Review TR100 as one of the top 100 innovators in the world under the age of 35. While at JPL, she founded and ran the Pasadena Delta Academy, a mentoring program for at-risk teen girls focused on STEM education, and the JPL Undergraduate Mentoring Program for Women, which provided mentoring support to undergraduate women engineering students.
A faculty member at Georgia Tech since 2005, Howard directs the Human-Automation Systems (HumAnS) Lab, where she blends control theory, human-robot interaction methods, and learning to the area of field and service robotics. She is the faculty lead of the I-Natural Vertically Integrated Project (VIP) Team, a multi-year multidisciplinary research team of undergraduates tasked to design, build, and test interfaces that enable humans to naturally interact with robots in performing activities of daily living. She has also teamed with faculty at the Atlanta Girls School to introduce a similar experience for a group of its 11th and 12th grade students.
In 2008, Howard received worldwide attention for her SnoMote robots, designed to study the impact of global warming on the Antarctic iceshelfs. She started a live blog to document the last glacier field trails in 2010 as part of a virtual scientist live videoconference with three minority-serving Boston high schools.
Over the last five years in Atlanta, Howard has hosted middle and high school robotics camps that have involved over 350 minority and female students and students with disabilities. At these camps, students were asked to solve challenges ranging from designing their own NASA lunar colony and robot assistants to programming AIBO robotic dogs to programming a robot avatar using a gaming scenario. She has also developed three educational software packages that allow K-12 teachers to adopt her hands-on learning initiatives.
Howard will join five other women receiving ABIE awards at the Grace Hopper Celebration of Women in Computing. In addition to Howard’s education award, these women will be honored for their achievements in technical leadership, social impact, international change agent, and emerging faculty member leadership.
The Anita Borg Institute (ABI), a non-profit organization focused on advancing women in computing, hosts the annual Grace Hopper Celebration of Women in Computing. This event is now the world’s largest gathering of women technologists, featuring keynotes by prominent women in technology, career workshops, and technical tracks from leading researchers.
]]>School of Electrical and Computer Engineering
404-894-2906
]]>The program aims to give students an understanding of a diversity of robotics topics, such as mechanics, interaction, perception, and artificial intelligence and cognition.
]]>The RoboCup “players” range in design from life-size humanoids to wheeled robots roughly the size of soccer balls. They will compete in five divisions on miniature indoor fields. The tournament runs from July 21-24, with the preceding weekend allotted for setup, testing, and an opening ceremony on Sunday. The event concludes on July 25 with an all-day symposium for adult participants and a workshop on educational robotics for RoboCup junior leagues.
Georgia Tech plans to compete in the Soccer Small Size League tournament using small, boxy robots with omnidirectional wheels. RoboJackets Vice President Lindsey Langstaff, an undergraduate student majoring in mechanical engineering, says she looks forward to working with other teams to help further research.
“You want to be able to bring something to the table that nobody’s come up with yet,” she says.
Organizers make the game tougher each year by changing parameters like field size or number of players. Next year, RoboCup co-chairman Alexandre da Silva Simoes, says, the robots might play outside.
RoboJackets is a diverse group of students, faculty, and alumni that aims to enhance the understanding of the field of robotics and its applications and increase the number of students exposed to robotics. They compete annually in five different robotics tournaments, serve as tour guides during the Institute for Robotics and Intelligent Machines’ annual National Robotics Week open house, and coordinate the regional FIRST Robotics Competition annual kickoff event.
]]>With RFE, however, patients get a mental boost. They are asked to think about moving. At the same time, a practitioner flexes the wrist. The goal is to send a long latency response from the stretch that arrives in the brain at the exact time the thought happens, creating a neural signal. The result is a strong, combined response that zips back to the forearm muscles and moves the wrist.
It all happens in a span of approximately 40 to 60 milliseconds.
“Timing is everything. When the window is that small, it’s not easy for two people to match each other,” said Georgia Institute of Technology master’s graduate Lauren Lacey.
That’s why Lacey and a team of fellow Georgia Tech researchers created a mechanical device that takes people out of the process, replacing them with accurate computers. Their functional MRI-compatible hemiparesis rehab device creates a long latency stretch reflex at the exact time as a brain signal.
“It’s kind of like trying to fill a bucket with water,” explained Minoru Shinohara, an associate professor in the School of Applied Physiology and director of the Human Neuromuscular Physiology Lab. “Stroke individuals can only mentally fill it halfway. The machine pours in the rest to make it full.”
So far, the research team has worked only with healthy individuals in their study. Study participants lie on a bed with the arm extended beneath a pneumatic actuator tendon hammer. In order to simulate the weak signal created by hemiparesis individuals to move their wrist, a transcranial magnetic stimulator (TMS) is placed on the heads of these healthy individuals at a 45-degree angle. Milliseconds after the hammer taps the wrist’s tendon, the TMS creates a weak signal in the motor cortex. The responses overlap, produce and send a strong signal back to the arm, and the wrist moves.
The team has successfully varied the timing of the TMS signal and speed of the hammer to strike faster or slower depending on how much of a boost is needed to complement the brain signal. Now that the researchers have proven the viability of the TMS-actuator system, they will next work with stroke individuals.
“The device is designed to adapt to people whether they are hyper, normo or hyporeflexive,” said Lacey, who graduated in spring with a master’s degree from the George Woodruff School of Mechanical Engineering.
Also, because the machine is MRI-compatible, it will allow the team to study what is happening in the brain during rehab, opening the door for robotics.
“Once we fully understand what is happening mentally and physiologically, we should be able to create a robot that can reproduce successful rehabilitative exercises such as RFE,” said Jun Ueda, an associate professor in the School of Mechanical Engineering. “It appears that the timing is the critical piece of this exercise. Robots are great at timing, so the results are very promising for robotics.”
The Georgia Tech team was assisted by researchers at Japan’s Kagoshima University, Kazumi Kawahira, Megumi Shimodozono and Yong Yu, who originally performed clinical studies of conventional RFE. The device was presented at the Design of Medical Devices Conference in Minneapolis, Minnesota this spring.
This research was partially supported by the National Science Foundation (NSF) under sub-award EEC 0540834. Any conclusions expressed are those of the principal investigator and may not necessarily represent the official views of the NSF.
]]>]]>
The researchers have paired a small humanoid robot with an Android tablet. Kids teach it how to play Angry Birds, dragging their finger on the tablet to whiz the bird across the screen. In the meantime, the robot watches what happens and records “snapshots” in its memory. The machine notices where fingers start and stop, and how the objects on the screen move according to each other, while constantly keeping an eye on the score to check for signs of success.
When it’s the robot’s turn, it mimics the child’s movements and plays the game. If the bird is a dud and doesn’t cause any damage, the robot shakes its head in disappointment. If the building topples and points increase, the eyes light up and the machine celebrates with a happy sound and dance.
“The robot is able to learn by watching because it knows how interaction with a tablet app is supposed to work,” said Georgia Tech’s Ayanna Howard, Motorola Foundation Professor in the School of Electrical and Computer Engineering who is leading the project. “It recognizes that a person touched here and ended there, then deciphers the information that is important and relevant to its progress.”
The robot analyzes the new information and provides appropriate social responses while changing its play strategy.
“One way to get robots more quickly into society is to design them to be flexible for end users,” said Hae Won Park, Howard’s postdoctoral fellow working closely on the project. “If a robot is only trained to perform a specific set of tasks and not able to learn and adapt to its owner or surroundings, its usefulness can become extremely limited.”
That flexibility is one reason Howard and Park see their robot-smart tablet system as a future rehabilitation tool for children with cognitive and motor-skill disabilities. A clinician could program the robot to cater to a child’s needs, such as turn taking or hand-eye coordination tasks, and then send the machine home.
Another benefit for rehab: parents don’t always have time or enough patience for repetitive rehabilitation sessions. But a robot never gets tired or bored.
“Imagine that a child’s rehab requires a hundred arm movements to improve precise hand-coordination movements,” said Howard. “He or she must touch and swipe the tablet repeatedly, something that can be boring and monotonous after a while. But if a robotic friend needs help with the game, the child is more likely to take the time to teach it, even if it requires repeating the same instructions over and over again. The person’s desire to help their ‘friend’ can turn a five-minute, bland exercise into a 30-minute session they enjoy.”
In a new study, Howard and Park asked grade-school children to play Angry Birds with an adult watching nearby. Afterwards, the kids were asked to teach a robot how to play the game. The children spent an average of nine minutes with the game as the adult watched. They played nearly three times as long (26.5 minutes) with the robot. They also interacted considerably more with the robot than the person. Only 7 percent of their session with the adult included eye contact, gestures and talking. It was nearly 40 percent with the robot.
The next steps for the Georgia Tech team will include more games for the robot, including Candy Crush and ZyroSky. They will also recruit more children diagnosed with Autism Spectrum Disorder (ASD) and children with motor impairments to interact with the system. Their most recent study included two kids with ASD. Their interaction times with the adult were significantly less than those in the typically developing group. They were about the same with the robot. The findings were presented in June at the Rehabilitation Engineering and Assistive Technology Society of North America (RESNA) 2014 Annual Conference in Denver.
This research was partially supported by the National Science Foundation (NSF) under grant 1208287. Any conclusions expressed are those of the principal investigator and may not necessarily represent the official views of the NSF.
]]>Got an unmanned aerial vehicle (UAV) sensor payload in need of testing? Well, Georgia Tech is set to offer defense customers an experimental aircraft on which to place it—at a fraction of the cost it would take to integrate that same payload on a conventional UAV.
The new test bed is called the GTRI Airborne Unmanned Sensor System (GAUSS). “It gives us the ability to offer proof of principle tests to customers at a price that’s reasonable, at a schedule that’s reasonable,” says Mike Brinkmann, MS EE 91, principal research engineer for sensor packages for the Georgia Tech Research Institute (GTRI).
GAUSS is based on the Griffon Aerospace Outlaw ER test UAV, which Tech purchased from Griffon and subsequently modified. The test bed has a 16-foot wingspan and weighs about 140 pounds, with a 35-pound payload capacity. Under Georgia Tech’s authorization from the Federal Aviation Administration (FAA), GAUSS can operate at a maximum ceiling of 5,000 feet, but it is capable of flying higher.
Some of the modifications GTRI researchers made to the Outlaw ER are immediately apparent. “In particular, we put pods on the wings to carry the radar system and power supply, and we made some modifications internally,” says Mike Heiges, AE 85, MS AE 86, PhD AE 89, GTRI’s principal aircraft research engineer for the project.
To prove it can test a variety of sensors on GAUSS, GTRI is integrating three different systems. The first is a visual light camera, the second is an RF signal detection package; and the third is a four-channel, side-looking radar designed to map the ground.
The radar is one of the first systems with these capabilities designed to be fitted on an aircraft as small as the GAUSS, and should be flying onboard it soon. “The two sensors that we have—the signals recorder and also the radar—we’re hoping will open some doors for GTRI to conduct sponsored research with a number of customers that would like to have combinations or variations on those things,” Brinkmann says.
Heiges adds that GRTI has an advantage over potential competitors because the Institute has authorizations from the FAA to allow it to fly the GAUSS at several locations around the country.
“That’s a huge deal,” Brinkmann says.
]]>Can studying the mating behavior of birds help the U.S. military develop better unmanned systems? That’s what Ronald Arkin, a roboticist at Georgia Tech’s College of Computing, and other researchers aim to find out as part of the U.S. Navy-funded Heterogeneous Unmanned Networked Teams (HUNT) Project.
Initiated in 2008, the HUNT Project is a multi-phased study that looks at assorted animal interactions—from wolves stalking an elk to squirrels hiding acorn caches—as inspiration for developing new algorithms to guide intelligent autonomous systems. For now, Arkin has been working with computer models and little bots in the lab. But things can always scale up to larger, more robust unmanned vehicles.
“That’s the beauty of the basic research,” he says. “It’s not limited to a physical type of platform.”
One of the earliest subjects of HUNT was “lekking” behavior in birds, in which a group of males gathers around—but not too closely—a very handsome specimen (a “hotshot”) in order to mate with females. This became the basis for seeing how one could distribute autonomous systems behind enemy lines “without using strict formation control” but in a way that “maximizes the likelihood of encounter” with the enemy, Arkin says.
In 2010 and 2011, Arkin and his team moved on to wolf packs. Initially, they thought the wolves coordinated with each other when hunting elk. But Dan MacNulty, a professor of wildlife ecology at Utah State University, disabused them of that notion. “When we brought Dan in the first time, he informed us that there is no coordination,” he says. “They are all individual, greedy agents.”
So how exactly did they work as a pack without explicit rules or communication? One possible explanation was that a predator chasing down an elk indicated to the others that the hunted animal was weak. So applying a probabilistic model to the stage of a hunt, Arkin tried to “replicate that behavior in robotic systems to see if we could do the same sort of thing both in simulations and platforms.” And he succeeded.
Following on the wolf pack research, Arkin then looked at bird mobbing, in which birds gather to drive off a stronger predator. Did it make sense for a weak bird to feign strength and participate in the mobbing? His simulations demonstrated that under certain conditions, yes, it did. And those same lessons could be applied to a low-power robot or one that’s out of ammo.
Arkin is now looking more broadly at robot deception. But, he explains, ultimately all of the pieces of HUNT relate to one another as examples of biologically inspired group behaviors.
]]>Despite the many famous animals that populate the annals of Georgia Tech lore—Sideways, Stumpy’s bear, the St. Bernards of Lambda Chi—for many years, the Institute rarely engaged with animals in an academic capacity. Even as recently as the late 1990s, every single research animal on Tech’s campus was contained in a single tank inside a lab in the School of Civil and Environmental Engineering. The sum total: six goldfish.
In the ensuing years, Georgia Tech researchers have expanded their focus to more fully explore the intersection of engineering and the natural world. And as they have, one theme has emerged again and again: For as much as we still have to learn about animals, they may have even more to teach us about ourselves.
These days, animals are helping researchers to better understand not only animals themselves but also the wider world, including humankind’s place within it—our physiology, our brains, our interactions with our environments. Animals are inspiring Georgia Tech’s faculty and students to create advanced robots, medical technology and improved prosthetics, among other developments that will shape the future, both saving and improving human lives.
Often, these animals aren’t the types that you would expect to be saving people. Take the sidewinder rattlesnake.
It’s best known for lying in wait in sandy stretches of the Southwest, ready to strike any prey that comes within reach and inject it with venom.
But the sidewinder’s unique motion that carves an arcing trail through the desert is proving key to researchers who seek to build a robot capable of moving across sand.
That is but one of a growing number of animal-related projects taking place on campus. As these endeavors have increased, Georgia Tech has taken steps to manage and oversee such work.
All research involving animals is conducted under strict guidelines ensuring that as few animals as possible are used in research, and that those animals are treated humanely. Animals have given much to researchers, and so researchers do their part to give something back.
Here, we look at just a small sampling of the ways in which animals are helping Georgia Tech researchers transform the world.
INSPIRING ROBOTS
Biologically inspired robotics has developed into a major focus at the Institute, with multiple labs looking to the animal kingdom for inspiration. One challenge that has long vexed researchers is the ability to traverse across sand. It’s a tricky prospect, as sandy surfaces can take on the properties of a solid, a liquid and even a gas.
But while robots struggle with the surface, various animals are able to move across sand, including lizards, sea turtles and snakes. Now Tech roboticists are mining the creatures’ behavior for their evolutionarily perfected secrets.
Animals are inspiring Tech’s faculty and students to create advanced robots, medical technology and improved prosthetics, among other developments that will shape the future, both saving and improving human lives.
A robotics team led by Dan Goldman, an assistant [associate] professor in the School of Physics, and David Hu, an assistant [associate] professor of mechanical engineering, began performing comparative studies on how sea turtles and sandfish (which essentially swim on land) move over sand.
Then they turned to snakes.
One snake-based robot that came out of the lab—known as Scalybot—was effective on many surfaces, but it always got stuck in sand. Many real snakes struggle with sand, too.
They partnered with Joe Mendelson, curator of herpetology at Zoo Atlanta and an adjunct professor at Tech, to study a snake that’s at ease on sand: the sidewinder rattlesnake.
“They’re famous for their funky sideways locomotion through sand dunes,” Mendelson says.
Georgia Tech prohibits venomous snakes on campus, and Tech researchers themselves can’t handle poisonous animals. Mendelson’s position at Zoo Atlanta allowed him to collect sidewinders from Arizona and conduct the research at the zoo in “the world’s most expensive sandbox.” Tech’s researchers simply observed the results.
The team now has a firm understanding of how sidewinders handle sandy slopes, and they’re examining how the snakes navigate obstructions. While sidewinder-style robots have obvious uses—search-and-rescue missions, military operations, planetary exploration—research partners from Harvard University have suggested sending the robots into sand-filled tunnels in Egyptian ruins.
“A robot can’t go down a sand-choked tunnel underground—only a snake can do that,” Mendelson says. “So we need a sidewinding robot with a camera that can look around. Then [if something of value is down there] you can put in the effort to dig it out.”
The needs of robots extend far beyond traversing sand, and inspiration has come from some surprising places. Hu received a lot of attention in 2012 for publishing a study of the “wet-dog shake”—when dogs shake wildly to dry themselves.
The physics of the wet-dog shake are impressive—dogs can shake themselves 70 percent dry in just a fraction of a second. While the research might seem silly, it does have useful implications. Hu says the research could be used for improved drying technology or in robotics.
“In the future, self-cleaning and self-drying may arise as an important capability for cameras and other equipment subject to wet or dusty conditions,” he says.
IMPROVING HUMAN LIVES
The School of Applied Physiology is home to the Comparative Neuromechanics Lab, where humans, rats and other creatures run on instrumented treadmills. Meanwhile, researchers gather data on how animals move.
This comparative data reveals a wealth of information on how healthy animals move, and how their bodies compensate after an injury. The findings are critical to the development of new approaches to rehabilitation of human and animal patients.
This research could be a potential alternative to bone-grafting operations.
The lab’s director, associate professor Young-Hui Chang, says he’d wanted to study animal locomotion ever since growing up watching animals in National Geographic documentaries.
The lab’s data also is being used in Tech’s Center for Prosthetic and Orthotic Research and Education to design and test new prosthetics, which are changing the lives of humans with missing limbs.
In the biotech quad on campus, the Parker H. Petit Institute for Bioengineering and Biosciences is focused largely on studying disease and injury and developing innovative treatments. One recent study showed that delivering stem cells on a polymer scaffold to treat large areas of missing bone led to improved results compared to using a scaffold alone. This research—conducted on rats—could be a potential alternative to bone-grafting operations.
“Massive bone injuries are among the most challenging problems that orthopedic surgeons face, and they are commonly seen as a result of accidents as well as in soldiers returning from war,” says the study’s lead author, Robert Guldberg, a professor of mechanical engineering and the Institute’s executive director. “This study shows that there is promise in treating these injuries by delivering stem cells to the injury site. These are injuries that would not heal without significant medical intervention.”
HELPING ANIMALS
Some researchers on campus have dedicated their time to developing models of animals. One team including researchers from applied physiology and biomedical engineering has developed a 3-D computer model that can be studied at almost the same level of detail as a physical specimen.
While the model can’t entirely replace live animals in experiments, it can greatly reduce the numbers that are used. The principal author of the model, Nathan Bunderson, also is in the process of making the model commercially available for educational purposes.
Another modeling effort that is providing a greater understanding of animals comes from the lab of Tech associate professor of physics Flavio Fenton.
Fenton has created extensive models of hearts after studying fish, mice and horses. His detailed electronic models are a tool to researchers and veterinarians around the world.
One project is focused on fostering a more symbiotic relationship between pets and humans.
The Tech-based Facilitating Interactions for Dogs with Occupations (FIDO) is an effort led by faculty member and dog lover Melody Jackson, PhD Computer Science.
Jackson, an associate professor of computer science and director of Tech’s Center for Biointerface Research, created a vest for canines that is equipped with several sensors.
A dog can trigger a sensor by nipping or nudging it, movements that send audible cues to the dog’s owner. The technology could be of use for service or rescue dogs.
“Currently, dogs can only communicate with people by barking or through body language. Sometimes that isn’t good enough,” Jackson says. “The sensors can give them a voice they’ve never had.”
The FIDO vest for canines is equipped with several sensors. A dog can trigger a sensor by nipping or nudging it, movement that send audible cues to the dog’s owner
The FIDO vest for canines is equipped with several sensors. A dog can trigger a sensor by nipping or nudging it, movements that send audible cues to the dog’s owner.
Caring for Research Animals
All Tech researchers whose work involves live specimens use as few animals as possible and follow strict regulations to ensure humane treatment.
These regulations are enforced by the Institutional Animal Care and Use Committee, a group that monitors all research and teaching activities at Georgia Tech involving vertebrate animals and makes certain it follows guidelines in the Federal Animal Welfare Act.
The IACUC reviews any activity involving animals before animals are used, and the committee meets monthly to review protocols. IACUC responsibilities include frequent inspections and documentation.
“At the deepest level, I owe animals the best possible care,” says Richard Nichols, professor and chair of the School of Applied Physiology at Tech and director of the Neurophysiology Lab, whose animal research augments his study of the physiology of human locomotion. “I feel particularly qualified to make sure of the humane treatment of my animals, and I regard it as a personal obligation.”
Learn more about the Georgia Tech Institutional Animal Care and Use Committee and the Institute’s policies regarding research animals.
Written by: Van Jensen
He recently made the startling prediction that robotics is advancing sufficiently such that children born today will never actually drive a car.
]]>Along with Chelsea Clinton, Girls Inc., Girl Scouts of the USA, Mindy Kaling, MIT Media Lab, National Center for Women & Information Technology, SevenTeen, TechCrunch and more, Google launched the new initiative on June 19. Since 2010, the company has invested $40 million to inspire girls to code, and Made with Code dedicates $50 million more over the next three years to further these efforts.
A member of Andrea Thomaz's Socially Intelligent Machines Lab, Fitzgerald focuses her research on human-robot interaction and cognitive science, where she codes robots to learn from people, much like how people learn from one another.
The Made with Code program includes:
Either way, the RoboJackets, a student group focused on robotics, are hard at work trying to bring this and other robo-feats to reality.
This weekend, a team of RoboJackets travels to Rochester, Michigan, for its first competition of the summer at the Intelligent Ground Vehicle Competition (IGVC). In this event, autonomous robots built by college students from around the world face off in an outdoor obstacle course.
“It’s a pretty tight-knit international community,” said Ben Nuttle, RoboJackets president and a computer science major. “We meet up and compete against a lot of the same people all the time.”
In IGVC, robots are designed to respond to obstacles in real-time, without any human interference. The RoboJackets employ stereoscopic cameras and several other technologies to get their robot through the course.
Later this summer, the RoboJackets will also send a team to RoboCup, where groups compete in varying leagues with robots designed to focus on specific robotics aspects or functions as they relate to playing soccer.
“We simplify the dynamics and mechanical challenges to really focus on things like intelligence, coordinating passing plays, and how you would strategically play a soccer game,” Nuttle said. “The emphasis of the competition is really on who has the best software.”
The RoboJackets also develop custom hardware from scratch — though the competition itself emphasizes software. Other behind-the-scenes work addresses hardware maintenance and upgrades, including the design and construction of custom control boards and new mechanical subsystems. At RoboCup, RoboJackets compete in the small-scale league, deploying small, omnidirectional robots that use custom radio signaling to play as a team. In the medium- and large-scale leagues, robots are closer to human size and focus on other aspects of mobility and intelligence. This year’s competition, which takes place July 17–29, serves as an encore to the actual World Cup, which takes place in Brazil June 12 – July 13.
While one RoboJackets team is in Brazil, another will be in Ontario from July 25–26, competing for the first time in the International Autonomous Robot Racing Challenge (IARRC).
The RoboJackets have been fielding robotics teams in various competitions since 1999, and in addition to participating in the three summer competitions, the group also hosts a FIRST Robotics Competition, where team members serve as mentors for high school students, and fields a team for BattleBots.
Around 10 students will travel to each competition this summer. The campus community can follow along via Facebook, and students interested in getting involved can contact Sue Pedapudi.
]]>
Kristen Bailey
Institute Communications
Arkin told The Wall Street Journal that systems like these should not be deployed unless they can comply with international humanitarian law. He added that the machines should be able to outperform human combatants from an ethical perspective, with the hope of leading to a reduction in civilian casualties in conflict. “If that bar cannot be met then they should not be deployed,” he said, adding that he supports a moratorium until such time as that bar can be met. “The original call for moratoria by the UN Special rapporteur were self imposed. I’m not sure if a ban or anything stronger is enforceable,” Arkin said.
]]>Before he became the first cyborg drummer, Jason Barnes was just a regular drummer, and like many musicians, he had a day job, a grubby grind cleaning the exhaust systems of restaurant oven.
So on a cool, misty January afternoon in 2012, Jason stood in a pool of water on the roof of a restaurant on McDonough’s historic square, working to pay the rent.
In a few days, he planned to audition to become a student at the Atlanta Institute of Music and Media. But his skins didn’t pay the bills.
He and assistant Nic Whisnant had already disassembled the fan covering the oven vent. The restaurant was closed, but the streets were full of cars joining the late afternoon rush hour.
Inside the kitchen the young men had rigged garbage cans to catch water and debris that flowed down as they scrubbed. Up on the roof they sprayed solvents down the exhaust shaft. Then Barnes, 22, picked up an aluminum pole with a magnetic pad on one end.
Outfitted with razor-sharp blades, the pad is designed to cling to the interior of an oven vent’s vertical shaft, making it easier to scrape off hardened crusts of grease inside as the worker pushes and pulls the pole from above.
This day was different. As Jason lifted the pole, he felt a sudden bolt of fear. “Get the (heck) down!” he screamed at Nic. In the same instant, he heard an explosion and saw a pink flash.
With his short, bristly blond hair perpetually standing on end, and his thin, wiry frame, Jason Barnes looks a little like a human lightning rod. That January day on the restaurant roof wouldn’t be the first or last time he’d attract a bolt from the hand of fate.
A few years ago he lost a significant chunk of his arm to an untreated spider bite.
Just recently he spun his mother’s Taurus on a wet highway, completing a Shaun White-worthy 540 before leaving the pavement. He and the car came out with a few scratches.
“He’s for sure a daredevil,” says friend Amy Ross, a tattoo artist who inked “L-I-V-E” on the knuckles of Jason’s right hand and “L-I-F-E” on the knuckles of his left. “Even if he’s not the one that says ‘let’s go do this,’ some of these things just find him.”
Jason could credit his mother Maggi Pier for his adventurous spirit.
Her life has been a vigorous and exploratory quest, moving around the world and experimenting with one business venture after another, not all of them successful. “I was always one to jump in the fire and then decide if it was hot,” says Pier, 63.
Jason’s father, who no longer lives with the family, is an Australian rock musician named Michael Barnes. Jason inherited that music gene and was playing in metal bands by the time he was a teenager, rotating between bass, guitar and drums.
He was born in Guam and spent part of his childhood in Roatán, a tiny island off the coast of Honduras in a cabin on the beach. He grew up skateboarding, racing motocross, cheating death and playing rock ’n’ roll. As a representative of the future of wearable robotics, Jason is an unlikely choice, having lived on the edge of civilization during his formative years.
But when he woke up in Grady Hospital, it soon became clear that technology would have to step in where flesh had failed.
He knew he’d been hurt, but he didn’t remember how; and he couldn’t figure out why his family was crying.
Then he looked at his hand.
“I was cooked pretty good,” is how Jason describes it.
His hair was singed and the right side of his face and right shoulder were bright red and blistered. Jason thought he’d been hurt in a fire or explosion; he had no memory of the accident.
Co-worker Nic would never forget it.
“The flash blinded me,” recalls Nic. They were both knocked off their feet. When Nic jumped up, he saw Jason lying on the roof. “All his hair was fried off, the sleeves to his clothes were frayed. He looked like a fish on a boat, gasping for air.”
His doctor estimates Jason had been hit with about 1,000 volts, after an arc of electricity crossed from an overhead power line into the pole he held in his right hand.
Surgeons sliced open his arm to relieve swelling and embedded pins in his fingers to keep them from curling into a permanent claw. Doctors also pulled strips of skin off Jason’s thigh and back to graft onto the damaged arm.
The efforts were in vain. His hand had suffered muscle and nerve damage. Most of its blood vessels were destroyed. After four or five operations it became clear he would never again use that hand to swing a drumstick again.
Jason faced one of the more painful decisions of his life. But to hear his family describe it, he was pretty matter-of-fact about it. Maybe he was just going stir crazy. Lying in a hospital bed is not his idea of a good time.
What’s the fastest way to get me out of here? he asked the Grady doctors.
Amputation, they said.
Looking back on the decision now, Jason counts the positives and discounts the negatives.
He points out that getting hot-wired could have turned out much worse. He could have been killed, or suffered brain damage, or lost his dominant hand. (He’s left-handed.) He had been crippled, yes, but he still had something crucial, spelled out in the letters on his left hand: L-I-F-E.
“I’m ridiculously lucky, if you think about it,” he says now.
He agreed to the operation, and then his pragmatic pluck deserted him. That night he grieved for all the things he would never get to do again and cried in his mother’s arms.
Days later surgeons removed Jason’s right arm below the elbow.
Jason left the hospital with a half-million dollars in medical bills and a small settlement from the restaurant that didn’t make much of a dent. He had to quit his job, give up his rental house in Jonesboro and move in with his mom near McDonough. Depressed and defeated, he went crazy with boredom.
“What I was going to do? Go home, veg out and play video games?” he remembers thinking. “No, you can’t do that. You can’t play drums anymore. You can’t play guitar anymore. My life was over. I was down and out.”
But then one day, three weeks after getting out of the hospital, Jason dragged his old drum set out of his mother’s garage. He still had bandages on his incisions and wore a silicone sleeve to protect the arm during physical therapy.
He took a roll of duct tape, attached a drumstick to his stump and tried out a simple pattern.
It was pathetic. And painful.
Playing hurt so much, it was impossible to do it for more than a minute. And the subtleties of drumming were erased by the crude arrangement. But he saw a light in that dark forest.
Things improved when a prosthetician at Hanger Inc. crafted a simple device that would hold a drumstick. Then Jason found a better custom prosthetic at TRS Inc., a company that makes appliances suited for archery, weightlifting, fishing, bicycling, guitar-playing and other activities.
He modified the device with his own hardware, using eyebolts and a spring from his kick drum. His goal was to create some “play” in the stick, a way for his prosthesis to mimic the “give” in a drummer’s grip.
He practiced with that appliance for about a year. “I can do this,” he thought. So he rescheduled the audition at Atlanta Institute of Music and Media he had missed the year before. In the fall of 2013 he was admitted to the school, where drum instructor Eric Sanders was impressed by Jason’s determination and persistence.
Nevertheless, there were some things Jason couldn’t do. A drummer’s grip can tighten or loosen, allowing the execution of single and double-stroke rolls. The more play in the stick, the more freely the stick bounces off the drum head. Without fingers, the grip stays the same.
Jason began to research myoelectrics, a type of prosthesis controlled by electrical impulses that are generated by the body’s muscles.
He dreamed of being able to use muscle tension in his right forearm to send signals to a mechanical hand, a hand that could instantaneously loosen or tighten its grip on a drumstick.
He told Sanders about his fantasy, but assumed it would never become a reality.
He didn’t know it already was.
Georgia Tech professor Gil Weinberg is a jazz pianist who speaks in rapid-fire bursts that make his Israeli accent even harder to decipher. The founding director of Tech’s Center for Music Technology, he likes to dream up bizarre machines and then build them. He’s the kind of futurist who demonstrates a technology breakthrough by jamming on a Miles Davis tune.
Weinberg has become famous in certain circles for developing robots that not only play music, but can improvise it. One of his creations is a marimba-playing android named Shimon that was featured on Stephen Colbert’s TV show. (Colbert joked that Shimon represented the two greatest threats to American life: jazz and robots.)
Why make robots that play music when we already have Skrillex?
The answer is that robots, just like 5-year-olds, benefit from musical training. The same algorithms that control Weinberg’s machines could help coordinate robot and human cooperation in other settings.
“We want robotic devices to anticipate what humans will be doing and synchronize their robotic actions just in time,” says Weinberg. “The idea is that if we get this to work in music — the most time-demanding medium — it would work in other scenarios, too. Think surgery rooms or space stations.”
Jason’s drum instructor Eric Sanders happened to see Shimon on “The Colbert Report” and found out Weinberg lived in his own back yard. He contacted the professor and told him about Jason, so Weinberg set up a meeting.
Jason politely smiled and rolled his eyes when Sanders told him about the professor.
Nothing will come of this, he said.
But while Jason imagined a device that could replicate a normal hand’s range of motions, Weinberg’s ideas went further. He wanted to create something superhuman, something that could play the drums like no one had ever played them before.
Weinberg secured a grant from the National Science Foundation to pursue the project and began to design Jason a $50,000 arm. He planned to debut it at the Atlanta Science Fest, less than eight months away, so the deadline was tight.
Using a light aluminum frame, Meka Robotics in San Francisco assembled a device with tiny, powerful, reversible electric motors to run belt-driven wheels that could swing a drumstick through a short arc. A second drumstick was embedded in the apparatus, a stick that Jason could deploy at will, but which would augment his playing with its own improvised additions.
Snapped into a carbon-fiber sleeve and hooked up to a nearby laptop, it’s an arm with a mind of its own.
“Jason?”
The name reverberated through the auditorium. Weinberg was announcing his new protege while an audience of 200 waited expectantly.
They had come to Kennesaw State University to hear the young cyborg drummer perform as the opening act of the inaugural Atlanta Science Fest.
But for a minute, Jason looked like he might be a no-show.
While Weinberg paged the drummer, Jason was pacing and smoking on a loading dock behind the Bailey Performance Center. He was trying to find a screwdriver to get his pincer prosthetic off. And he wasn’t happy: Weinberg had just told him he wanted him to stand up in front of the audience and tell his story.
Playing drums, even playing with just one hand, is much easier than public speaking.
“I don’t know what I’m going to say,” he fumed.
It had been a stressful month. Meka delivered the working arm at the end of February, less than a month before the Science Fest debut. Jason tried it on, played with it for about 30 minutes, and then Weinberg asked him to perform, right then and there, with a group of musicians for a promotional video. Terrified, Jason felt like a student driver being handed the keys to a Lamborghini. Now, once again, he was being forced out of his comfort zone.
Jason stepped out on the Kennesaw stage and gave the audience a concise, unemotional account of his harrowing experience.
Then Jason sat down at his black drum kit and immediately relaxed. He played a percussion duet with Sanders, and then a version of Miles Davis’ classic “So What?” in a combo featuring Weinberg at the piano.
Then Weinberg explained that he had downloaded some rhythmic patterns from atom bomb scientist (and drummer) Richard Feynman, and uploaded those beats into Jason’s arm. This allowed Jason to play a duet with a dead theoretical physicist.
Finally, Jason demonstrated some of the super-human qualities of the arm. It can play 20 beats a second — essentially a one-handed roll. It can play contrapuntal rhythms — say, five against eight. It’s an arm that sounds like two drummers playing at once. Speed metal drummers would be envious, Jason observed.
Though his game face didn’t offer a clue to his feelings, Jason was elated. After the show he pulled off the robot arm, donned his regular pincer, and posed for photographs with giddy audience members, who mobbed him in the lobby.
“Look at him,” said friend Nic later. “He’s bigger and badder than he ever was. He is a way better drummer, and a better person.”
On the wall in Jason’s practice shed is a portrait painted by his tattoo artist friend, Amy Ross. It shows Jason’s face, surrounded by lightning-bolt letters that spell out “Cheat Death.”
The painting helps him keep things in perspective when he gets frustrated. Like when he wants to change the station on the car radio, but has to use his left hand to do it. When he tries to hold his smartphone with his pincer and cracks the screen. When he feels a tiny little itch on the end of his right thumb, which he can’t scratch, because he doesn’t have a right thumb any more.
He is, after all, still here. “I shouldn’t be alive, honestly,” he says. “I mean depression sucks and everything, but feeling sorry for yourself is not going to get you anywhere.”
And even though it was a total hoot to become the first bionic drummer, the robot arm didn’t really solve all of his problems.
Jason talks about the drawbacks of a $50,000 arm as he shows off his music headquarters, a small outbuilding at his mother’s McDonough house, where the walls are hung with guitars and beer posters and assault rifles, and where his band, when he has one, gets together.
The tour is interrupted by a cellphone call. It’s another journalist requesting an interview, one of dozens he’d done in recent weeks. He’s been on CNN, Fox News, MSNBC, NPR, the Discovery Channel, a German news station and in a handful of print publications.
This is the biggest change: Jason has become a mini-celebrity. “He’s going to be in Wikipedia!” marveled Maggi. “I Googled his name: It’s page after page!”
To help capture some of that buzz, the family hired a videographer to document the Science Fest concert, and Jason has his own YouTube channel.
But will he continue to be the robo-drummer? The possibilities are tantalizing. Shortly after his Georgia Tech video went public, Jason was contacted by Rick Allen, the drummer from the rock band Def Leppard, whose left arm was severed in a street-racing accident. Allen invited Jason to meet him backstage when his band plays Atlanta in July. One can’t help but imagine a duet between the two one-handed drummers, each aided by high-tech devices.
The fact is, the expensive robo-arm does not belong to Jason, it belongs to Georgia Tech. And while he is grateful to be part of Weinberg’s experiment, it’s clear that Jason prefers his old drum-hand prosthetic. It may not have a mind of its own, but is lighter and more flexible. The Meka arm weighs almost two pounds, which is like holding a quart of beer at arm’s length while trying to play music.
Yet the recognition he’s gained may be enough to help Jason rise above the sea of other struggling musicians. The life of a musician is a challenge; for drummers the sailing is even rougher, considering the popular taste for electronic dance music and computerized beats.
Jason’s high profile has already paid off in a few ways. He recently traveled to Los Angeles with his mother for an appearance on the TV show “The Doctors,” where the producers promised him a bebionic mechanical hand, a $60,000 myoelectric prosthetic device.
Maggi Pier is convinced the best is yet to come.
“I’ve known in my heart of hearts that Jason will become something big because of this,” she says, sitting on her front porch looking out over her rural property. “It’s funny how God works, how you have to go through fire to get to something better.
“You can’t let having your arm chopped off keep you from getting on with your life.”
HOW WE GOT THE STORY
I first met Jason Barnes at his prosthetician’s office, where he was looking for a new hand. I was going to feature him in a story about the Atlanta Science Fest. Among the dozens of technological marvels revealed at the festival, his robot arm was the coolest. But after I heard his story, I realized it was a deeper, more complex story better suited to a Personal Journey. Later I met his mother, siblings and friends, heard him perform in concert and had a chance to see some of his stranger tattoos up close. (His leg-tat portrait of Michonne from “The Walking Dead” is creepy and impressive.) Jason was just trying to do his job when a freak accident shoved him into the spotlight. He didn’t ask for the amputation, or the attention, but he has handled both with grace.
Bo Emerson
Staff writer
personaljourneys@ajc.com
Bo Emerson is an Atlanta native who joined The Atlanta Journal-Constitution in 1983. He has been a feature writer for most of his AJC career, covering music, the Olympics and Billy Graham’s last crusade. He is also a musician and plays jazz trumpet with Style Points, The Lowlights, and other hackers. He is married to Maureen Downey, who covers education for the AJC.
David Tulis is an Atlanta photojournalist who spent most of his career at The Atlanta Journal-Constitution before striking out on his own in 2009. He has covered the Olympics, the World Series-winning Atlanta Braves team, and traveled to South America, Europe and Africa for the newspaper. He is a member of Georgia State University’s 100th Anniversary Class of 2013 and plays bass guitar with the Sagamores.
]]>Stilman, who was recently awarded tenure and scheduled to be promoted to associate professor effective Aug. 16, joined the Georgia Tech faculty in 2008 after receiving his Ph.D. in robotics from the Robotics Institute at Carnegie Mellon University.
“We are stunned by this news about our colleague and friend Mike,” said Annie Antón, professor and chair of the School of Interactive Computing. “His promise as a brilliant young researcher was shown this year by his promotion to a tenured position and his winning of a prestigious institute-wide teaching award. This is a loss from which we will not easily recover.”
An active member of the Institute for Robotics and Intelligent Machines (IRIM) at Tech, Stilman founded the Humanoid Robotics Lab, also known as Golems.org, where he led a team of graduate students and researchers on projects designed to develop “robot brains,” or algorithms, capable of the same level of dexterity, intelligent planning and interaction as human brains.
“Mike was a pioneer on motion planning and humanoids,” said Henrik Christensen, executive director of IRIM. “He was breaking new ground in his research, but he was also an eminent teacher. His energy was contagious, and he was always ready to push us even further. He will be sorely missed as part of the robotics and broader community at Georgia Tech.”
As an assistant professor in the School of Interactive Computing, Stilman oversaw the research projects of 16 graduate students and several undergraduates. Recognized by students as a favorite in the classroom, he taught several robotics courses, including Robot Intelligence and Humanoid Robotics, and earned the Class of 1940 W. Roane Beard Outstanding Teacher Award for the past academic year.
Known for his enthusiasm for humanoids (his Georgia license plate reads “GOLEM”) as well as his abundance of energy, Stilman was active in the broader robotics research community. Most recently, he chaired the IEE-RAS International Conference on Humanoid Robotics in Atlanta and led Tech’s participation in the DARPA Robotics Challenge.
Stilman was among 25 researchers nationwide selected for a 2013 Young Faculty Award by the Defense Advanced Research Project Agency (DARPA). Also, he received several significant research contracts involving his two humanoid robots, Golem Krang and Golem Hubo, including a $900,000 grant from the Office of Naval Research to develop a “hybrid reasoning system” for military robots. The new system could lead to robots that improvise, using objects in their environments to creatively complete complex tasks, such as climbing a chair to escape a building. In a 2012 interview about this project, Stilman said his goal was “to create machines that operate like MacGyver, the ‘80s TV character.” This quote marked the birth of Golem Krang’s new moniker: “MacGyverBot.”
“This is a shocking loss for the College of Computing and the entire Georgia Tech robotics community,” said Zvi Galil, dean of the College of Computing. “Mike Stilman was a brilliant and passionate young teacher and researcher with an outstanding career in front of him. On behalf of everyone in our college, I extend our deepest condolences to Mike’s family. His loss will be felt within our community for a long time.”
Survivors include his wife Akiko and his parents.
]]>The Robotic Industries Association, a U.S. trade group, last week organized its first conference dedicated to collaborative robots, at which robot manufacturers and customers gathered to discuss the trend. Christensen was a keynote speaker.
]]>“My current prediction is that kids born today will never have to drive a car,” he said in this interview with FW:Thinking. It’s a reasonable conclusion — consider Google’s self-driving car system, which is already street-legal in Nevada, Florida, Michigan, and California. (Texas has similar legislation already in the works.)
]]>For the 2014 National Robotics Week, Georgia Tech's Institute for Robotics & Intelligent Machines (IRIM) has partnered with IEEE Spectrum and iRobot to deubt twice the robotic talent, with two brand new decks of trading cards.
IRIM has designed a fresh pack of Georgia Tech Team Cards, highlighting a new lineup of innovative automatons. In addition to the 2014 Georgia Tech cards, IRIM teamed up with National Robotics Week coordinators iRobot and IEEE Spectrum to create a "National Team" featuring famous robots from around the country--including IRIM's own Simon, created by Associate Professor Andrea Thomaz of the School of Interactive Computing. To see both sets, plus the original 2013 lineup of Georgia Tech robots, visit the IRIM Robot Trading Cards webpage.
Professor Henrik Christensen, KUKA Chair of Robotics at the Georgia Tech College of Computing, touts the robot trading cards as an earmark for the extraordinary future of robotics.
"We're thrilled not only to put out our second 'lineup' of robotic All-Stars but also to work with iRobot and IEEE to promote a 'National Team' of automated heavy hitters from around the country," says Christensen. "This is a terrific way to help future roboticists learn about and appreciate all the fascinating work going on in U.S. robotics. Perhaps, 50 years from now, scientists will point back to their vintage Curi rookie cards as the spark for a lifetime of creative discovery in this critical, wide-ranging field."
The original cards were created as an avenue through which the public could connect with this dynamic and exploding field of science and engineering.
"One thing I learned covering the field of robotics is that people really love robots," says Erico Guizzo, the robotics editor at IEEE Spectrum and one of the creators of the IEEE Robots for iPad app. "People love robot videos, robot articles, robot slideshows, and now they'll love the robot cards."
Though National Robotics Week is meant as a showcase for new robotic talent, Lisa Freed, STEM Manager for iRobot, emphasizes that education is also a primary goal of this celebration.
“We at iRobot are thrilled to have worked with Georgia Tech and IEEE on this set of exciting robot trading cards," Freed says. "Our hope is that these will help to further engage young students by showcasing the diverse world of robots in a new way.”
IRIM will distribute free packs of the 2014 cards to visitors touring Georgia Tech's robotics labs on April 8, as the campus plays host to hundreds of middle- and high school students visiting as part of National Robotics Week activities. Tonight, a limited number of cards will be handed out at Russ Chandler Stadium when the Yellow Jackets men's baseball team takes on Georgia Southern at 6 p.m.
For more information, visit the National Robotics Week at IRIM webpage. As a special surprise for the collectors of this year's cards, IRIM has designed a digital team card featuring the whole Georgia Tech lineup, which can be downloaded by liking IRIM's Facebook page.
]]>Digital Content Specialist
404.385.3870
]]>In a proof-of-concept study, scientists selected a region for analysis on round or irregularly-shaped objects using a 3-D camera on a robotic arm, which mapped the 3-dimentional coordinates of the sample’s surface. The scientists programmed the robotic arm to poke the sample with an acupuncture needle. The needle collected a small amount of material that the robot deposited in a nearby mass spectrometer, which is a powerful tool for determining a substance’s chemical composition.
“You see the object on a monitor and then you can point and click and take a sample from a particular spot and the robot will go there,” said Facundo Fernandez, a professor in the School of Chemistry and Biochemistry, whose lab led the study. “We’re using an acupuncture needle that will touch very carefully on the surface of the object and then the robot will turn around and put the material inside of a high resolution mass spectrometer.”
The research was published online February 28 in the journal Analyst, a publication of the Royal Society of Chemistry. The research will be featured on the cover of an upcoming print issue. The work was supported by a National Science Foundation (NSF) Major Research Instrumentation Program (MRI) grant and by the National Science Foundation (NSF) and NASA Astrobiology Program, under the NSF Center for Chemical Evolution.
Mass spectrometry is a powerful tool for analyzing surface chemistry or for identifying biological samples. It’s widely used in research labs across many disciplines, but samples for analysis typically have to be cleaned, carefully prepared, and in the case of rocks, cut into thin, flat samples. The new robotic system is the first report of a 3-D mass spectrometry native surface imaging experiment.
“Other people have used an acupuncture needle to poke a sample and then put that in mass spec, but nobody has tried to do a systematic, three-dimensional surface experiment,” Fernandez said. “We are trying to push the limits.”
To show that the system was capable of probing a three-dimensional object, the researchers imprinted ink patterns on the surfaces of polystyrene spheres. The team then used the robotic arm to model the surfaces, probe specific regions, and see if samples collected were sufficient for mass spectrometry analysis. The researchers were able to detect inks of different colors and create a 3-D image of the object with sufficient sensitivity for their proof-of-principle setup, Fernandez said.
The research was the result of collaboration between Fernandez’s group, which specializes in mass spectrometry, and Henrik Christensen’s robotics group in the College of Computing. Christensen is the KUKA Chair of Robotics and a Distinguished Professor of Computing. He is also the executive director of the Institute for Robotics and Intelligent Machines (IRIM) at Georgia Tech.
“The initial findings of this study mark a significant step toward using robots for three-dimensional surface experiments on geological material,” Christensen said. “We are using the repeatability and accuracy of robots to achieve new capabilities that have numerous applications in biomedical areas such as dermatology.”
“It doesn’t happen very often that a group in mass spectrometry will have a very talented robotics group next to them,” Fernandez said. “If we tried to learn the robotics on our own it could take us a decade, but for them it’s something that’s not that difficult.”
Christensen’s team loaned a Kuka KR5 sixx R650 robot to Fernandez’s lab for the study. Afterwards, Fernandez’s lab purchased their own robot from Universal Robots. They have also upgraded to a new mass spectrometer capable of resolution nearly eight times higher than the one used in the study. They will soon begin replicating early Earth chemistry on rocks and analyzing the reaction products with their robotic sampling system.
“We really want to look at rocks,” Fernandez said. “We want to do reactions on rocks and granites and meteorites and then see what can be produced on the surface.”
The technology could also be applied to other research fields, Fernandez said. For example, the robot-mass spec combo might be useful to dermatologists who often probe lesions on the skin, which have distinct molecular signatures depending on if the lesion is a tumor or normal skin tissue.
This research is supported by the American Recovery and Reinvestment Act (ARRA) under the National Science Foundation (NSF) Major Research Instrumentation Program (MRI) (Grant number 0923179), and by the NSF and NASA Astrobiology Program under the NSF Center for Chemical Evolution (CHE-1004579). Any conclusions or opinions are those of the authors and do not necessarily represent the official views of the sponsoring agencies.
CITATION: Rachel V. Bennett, et al., “Robotic Plasma Probe Ionization Mass Spectrometry (RoPPI-MS) of Non-Planar Surfaces.” (Analyst, February 2014) http://dx.doi.org/10.1039/c4an00277f
Research News
Georgia Institute of Technology
177 North Avenue
Atlanta, Georgia 30332-0181 USA
@GTResearchNews
Media Relations Contacts: Brett Israel (@btiatl) (404-385-1933) (brett.israel@comm.gatech.edu) or John Toon (404-894-6986) (jtoon@gatech.edu)
Writer: Brett Israel
]]>404-385-1933
]]>“By discovering a principle behind this ability — localized fluidization — the researchers are able to give a robotic digging clam, RoboClam, similar abilities. And the study of the robot gives deeper insight into the important mechanics behind burrowing through localized fluidization,” he adds.
]]>“The second drumstick has a mind of its own,” said Weinberg, founding director of the Georgia Tech Center for Music Technology. “The drummer essentially becomes a cyborg. It’s interesting to see him playing and improvising with part of his arm that he doesn’t totally control.”
The prosthesis was created for Jason Barnes, a drummer who was electrocuted two years ago and lost his right arm below the elbow. The Atlanta Institute of Music and Media student built his own prosthetic device shortly after the accident. It wasn’t very flexible. He could bang the drums by moving his elbow up and down, but couldn’t control the speed or bounce of the stick without a wrist or fingers. That’s when Weinberg stepped in to create a single-stick device with sensors that responds to Barnes’ bicep muscles.
“Now I can flex and send signals to a computer that tightens or loosens the stick and controls the rebound,” said Barnes.
Weinberg, who has already built a robotic percussionist and marimba player that use computer algorithms to improvise with human musicians, took the prosthesis a step further. He added the second stick and gave it a “musical brain.”
“Jason can pull the robotic stick away from the drum when he wants to be fully in control,” says Weinberg. “Or he can allow it to play on its own and be surprised and inspired by his own arm responding to his drumming.”
Regardless of how he uses the extra stick, the new prosthetic has already given Barnes capabilities he hasn’t had since before the amputation. It’s only the beginning for Weinberg. He’s using a National Science Foundation grant (IIS- 1345006) to expand the technology.
“Music is very time sensitive. You can hear the difference between two strokes, even if they are a few milliseconds apart,” said Weinberg. “If we are able to use machine learning from Jason’s muscles (and in future steps, from his brain activity) to determine when he intends to drum and have the stick hit at that moment, both arms can be synchronized.”
Weinberg says such robotic synchronization technology could potentially be used in the future by fully abled humans to control an embedded, mechanical third arm during time-sensitive operations. For example, Weinberg’s anticipation algorithms could be used to help astronauts or surgeons perform complex, physical tasks in synchronization with robotic devices.
For Barnes, it’s all about the music. Because an embedded chip can control the speed of the drumsticks, the prosthesis can be programmed to play two sticks at a different rhythm. It can also move the sticks faster than humanly possible.
“I’ll bet a lot of metal drummers might be jealous of what I can do now,” he said. “Speed is good. Faster is always better.”
Barnes will play with the device for the first time publicly on March 22 at the Robotic Musicianship Demonstration and Concert at Kennesaw State University’s Bailey Performance Center. The free event, which will also feature Weinberg’s Shimi robot, is part of the Atlanta Science Festival.
]]>The advances of undergraduate Jamie Clark, alumna Melissa McCoy, and faculty member Karen Feigh offer a glimpse of what Georgia Tech’s women in engineering are capable of accomplishing.
Read the full story
Women in Engineering at Tech >
He was sweating it out.
Nearly two decades ago—before search engines or smartphones—Thad Starner, then 24, sat in front of a qualifying panel at the Massachusetts Institute of Technology. For several hours he was grilled by four professors, a prerequisite before he could write his Ph.D thesis.
At the end, one of his inquisitors asked: Hey, Thad, were you using your wearable computer?
Starner, clad in a get-up that covered his left eye with wires leading down to a keyboard on his forearm, plus a satchel holding a battery pack and a measly 80 megabyte hard drive, said: Yes.
His answer touched off a half-hour debate that almost culminated in MIT conferring its degree on “Thad Starner and his computer.” That, sadly, did not come to pass.
“It would have been hilarious,” the 44-year-old, now on faculty at Georgia Tech, said in hindsight.
Starner’s work at MIT’s Media Lab would later lay some of the groundwork for Google Glass. Since 2010, he been a technical lead for the project, as well as the founder and director of the Contextual Computing Group at Georgia Tech’s College of Computing.
He was one of the people to wear the first iteration of Google’s wearable computer — which he found a distinct improvement on his earlier model. (He did, though, have to cut his shoulder-length hair in order to wear it comfortably.)
He now wears the latest version, in light blue, fitted with his prescription lenses.
This weekend Google is bringing the circus to town, inviting regular folks to try on Glass at the Foundry at Puritan Mill.
The two day event is the fourth stop on the search engine’s campaign to make people more comfortable with the $1,500 devices. Google hopes the tour will help dismiss some of the Orwellian perceptions that have surrounded the device over the past year.
Glass wearers look up into a translucent heads-up display (not unlike that of a fighter pilot), where information is displayed. The device is based on more than a half-century’s worth of research, dating back to 1945.
Google would like you to think about Glass like this:
The cloud is forever. Your laptop is a place where you can hold information for months or years. Your smartphone is for stuff you want to keep for a week.
But wearable computing, an emerging category that includes Glass, is for right now — the Random Access Memory (RAM) of our lives.
Glass—in fact all wearable computing—is meant to function almost as an extension of the wearer’s nervous system. In effect, Starner is a beta site: a trial run for of what a computer-enhanced human might look like.
He’s been taking computerized notes on nearly every important conversation he’s had since at least 1993; asked about something from his past, he uses Glass to retrieve the notes.
(Wearers use a combination of eye and head movements and spoken commands to access information and perform other operations on the Glass computer.)
The stories Starner tells about the people he’s met and places he’s been come out in encyclopedic bursts. No chronology. Just search-engine-like depth and speed. Jumping from one subject in 1995 to another in 2008.
In the midst of talking to you, he looks up, responding to an incoming alert. He stares into space, reading material returned in his latest web search. His eyes flick upward; he’s gotten an email. You get the gist.
It sounds strange, but it’s actually meant to be less intrusive than having the person you’re talking to constantly dive into their cell phone to handle the same range of tasks.
On Thursday night, during a last-minute Google Glass-sponsored Design Sprint on the sixth floor of the Centergy Building in Midtown, four Google development specialists chatted about the ins-and-outs of creating software for Glass.
The gathering drew roughly a 100 Georgia Tech students, programmers and designers (some cool enough to wear both Glass and green pants).
That kind of buy-in from the development community is critical, especially if Glass is to take off. No one company has the ability to fill the needs of all mobile Internet users.
To hear the Google folks tell it, Glass is almost like a silent librarian.
It shouldn’t interfere with the actions of the user. Glassware (as Google calls its brand of wearable computing apps) should only give gentle reminders and directions.
To refine the user experience, Google has been testing its foray into wearable computing for the past year on Glass Explorers.
That’s a group of early adopters who were originally chosen from a pool of roughly 150,000 applicants. Each explorer paid $1,500 for the privilege of being a test subject, a price point that is unlikely to fall anytime soon.
Their ranks have grown from 8,000 at the onset to roughly 10,000.
“‘When you concentrate on the hardware, the software you get creates bad interactions,” said Starner, sitting toward the back of the room. He wore a Mandarin-style shirt buttoned to the top, a dark sports coat and black jeans.
“And when you start with the human and design the software around the person, it works very well.”
Sitting on a couch at the edge of the space, a millennial — a digital native, maybe a Google employee, herself — stared up into the bifurcated Glass screen, which might have been displaying an important text message, just out of her normal line of vision.
After a moment, she gracefully shifted her attention back to the presentation, but then a real interruption intruded. She darted from the room to answer a smartphone call.
Glass is set to be released to the masses later this year.
]]>This article is one in a series of Q&As to introduce the Tech community to the 10 IRIs and their leaders. In this installment, Executive Director of the Institute for Robotics and Intelligent Machines (IRIM) Henrik Christensen answers questions about IRIM and also talks about its efforts to support Georgia Tech faculty and students.
Q: What is the Institute for Robotics and Intelligent Machines (IRIM), and what are its core research areas?
A: The Institute for Robotics and Intelligent Machines is a new IRI that integrates robotics research, education and outreach, and industry engagement across the College of Engineering, the College of Computing, the College of Sciences, and the Georgia Tech Research Institute (GTRI). Our work often involves labs and individual researchers in other Georgia Tech colleges and centers, as well.
We conduct research in mechanisms, control, perception, artificial intelligence (AI), and human–robot interaction (HRI) with a particular emphasis on human-centered robotics. The question, “How can we build robots that empower people in their daily lives, whether for service in the workplace or in the home, or for enjoyment in a leisure setting?” is central to our work.
Using robots makes it possible to compete with low-wage manual labor in other countries. It also creates new positions that replace the dirty, dull, and dangerous jobs in U.S. factories. Additionally, robotics technologies have made it possible to improve the quality of life in an aging society by providing services that allow people to remain autonomous as they lose various functions such as mobility and memory. Finally, our research leads to new types of autonomous systems to assist first responders and soldiers during interventions by increasing the distance between responders and the immediate danger, including fires, earthquakes, and explosives.
IRIM has three objectives: 1) to be the world leader in human-centered robotics, 2) to educate the best people to serve in academia and industry for next-generation robotic systems, and 3) to create new opportunities in robotics for industry and society at large, in both Georgia and beyond.
Q: A lot seems to be going on in robotics these days. Can you summarize the big trends and Georgia Tech’s role with regard to those trends?
A: Robotics has seen tremendous growth in the past few years. Today, robots are used to re-shore jobs to the U.S. in industries such as automotive, aerospace, and electronics manufacturing. We have also seen the development of major new services for the home – from robot vacuum cleaners to autonomous transportation and personal assistance devices. And, of course, we have seen numerous robots used in Iraq and Afghanistan to make life a little safer for our soldiers.
Overall, we are seeing major growth in manufacturing, e-commerce, health care, and service industries.
The U.S. recently initiated a number of big programs in robotics, such as the National Robotics Initiative (NRI), which is sponsored jointly by the National Science Foundation, National Institutes of Health, the U.S. Department of Agriculture, and NASA. The NRI was launched on the basis of the Roadmap for U.S. Robotics, a report initially published in 2009 and revised in 2013. Georgia Tech served as the coordinator of the development of both editions of this report. To support the NRI, a national network, the Robotics Virtual Organization was founded and is managed by Tech. Consequently, Tech is seen, in many respects, as the leader for the push for new robotics initiatives in the U.S. across research, education, and the translation of results.
Q: How does IRIM support research?
A: IRIM supports the research of more than 60 faculty members and 140 graduate students across various colleges and GTRI in a number of ways.
First, we proactively identify major new funding areas and launch seed projects that allow Georgia Tech to be competitive when calls for proposals are issued. There are remarkably few opportunities for faculty to conduct exploratory research without funding constraints, so we try to identify these new opportunities early and build up results to ensure we can successfully compete for funds.
Additionally, we are developing an infrastructure that matches researchers with similar interests so, together, they have a more competitive edge when applying for major funding awards. Although our researchers are very good at pursuing grants, it is challenging, as a single applicant, to generate adequate support to build a successful proposal for major funding awards such as NSF’s Engineering Research Centers (ERCs) or Science and Technology Centers (STCs) grants. For example, it is difficult for one faculty member to build a complete manufacturing facility for new robotics research in the automotive industry. However, IRIM can provide a shared infrastructure that allows multiple researchers to pursue a larger research effort in a shared space.
IRIM is also committed to providing support to faculty pursuing major research opportunities through all phases of the process, from early research efforts and proposal writing to grant management and evaluation of broader impact and outreach. We would rather see our robotics faculty winning a smaller number of major grants rather than a larger number of smaller grants because comparatively, the smaller grants have too much overhead.
Additionally, IRIM facilitates opportunities for engagement in interdisciplinary activities through events such as weekly seminars and topical workshops throughout the fall and spring semesters.
Finally, our One Georgia Tech approach allows external stakeholders, especially our industry partners, the chance to work with IRIM to identify the individual or lab on campus that best matches their research needs.
Q: How is IRIM furthering Georgia Tech’s academic mission?
A: Over the past few years, we have built a strong Ph.D. program in robotics in which we currently have close to 50 graduate students enrolled. These students are required to have an interdisciplinary focus and must choose coursework that involves three of five core robotics areas: mechanics, controls, perception, HRI, and AI and autonomy. Our interdisciplinary approach has proven to be very popular with students, as well as with employers.
Additionally, IRIM is working on the development of a professional master’s program in robotics. Georgia has a strong industry base related to robotics, and many of these companies would welcome the opportunity to have a continuing education program available locally for their employees. A professional master’s program would not only allow us to attract more students to Georgia Tech, it would also build new links to industrial companies from across the state.
IRIM also actively engages with undergraduate students enrolled in participating units (Interactive Computing, Electrical & Computer Engineering, Mechanical Engineering, Biomedical Engineering, and Aerospace Engineering) through coursework and undergraduate research opportunities. This summer, we are launching an NSF-sponsored Summer Undergraduate Research in Engineering (SURE) Program for students to spend summer on campus to conduct research with robotics faculty and graduate students. We see this program as a strong recruiting mechanism to attract the best students to Georgia Tech for graduate studies.
Q: How does IRIM support industry engagement and community outreach?
A: IRIM has a proven track record of cultivating successful industry partnerships, including those with KUKA, Boeing, General Motors, BMW, PSA Peugeot Citroën, Google, Microsoft, iRobot, and Lockheed Martin.
Through a strong collaboration across academic units and GTRI, IRIM offers industry partners access to a broad research portfolio, as well as an abundance of beneficial services that span from basic research opportunities to full-product development solutions. Too often, innovations are lost in the abyss between basic research and applications. IRIM has the faculty, processes, and experience to ensure these innovative projects can be successful. Few other academic or research institutions in the U.S. have a comparable scope of expertise and options available to industry.
For broader community outreach, IRIM works closely with organizations across Georgia and the nation, such as high schools, to provide education on the impact of robotics with regard to everyday living. We do this through initiatives such as the FIRST Robotics Competition. The undergraduate robotics club, RoboJackets, with support from IRIM, organizes the annual kickoff for this competition. In 2013, more than 1,000 high school students attended the event at Ferst Center for the Arts, and quite a few Georgia Tech students and faculty members are mentors for the FIRST team.
Additionally, in an effort to stimulate general interest in STEM subjects, as well as a specific interest in robotics, IRIM organizes regular school visits across Georgia during the year. Since the launch of National Robotics Week in 2010, IRIM has participated annually by sponsoring an open house at Tech and conducting lab tours and demonstrations for middle and high school students. More than 400 students participated in Tech's 2013 event held on April 11, with one group traveling from Tennessee to attend. Tours offered participants a chance to learn more about 46 different research projects in 16 different robotics labs on campus. We anticipate the 2014 event will be even bigger and better than last year!
]]>
You’ve probably heard that Georgia Tech has a number of Interdisciplinary Research Institutes (IRIs) – but do you know much about them?
This article is one in a series of Q&As to introduce the Tech community to the 10 IRIs and their leaders. In this installment, Executive Director of the Institute for Robotics and Intelligent Machines (IRIM) Henrik Christensen answers questions about IRIM and also talks about its efforts to support Georgia Tech faculty and students.
]]>Director, Research Communications
]]>But after talking with more than a hundred potential users of the device, she learned the real need was for a generic interface system able to connect a wide range of input devices – big button switches, joysticks, sip-and-puff straws and others – to the tablet computers. And it turned out that the market was much larger than Howard imagined, extending to adults with disabilities and potentially even persons with Alzheimer’s.
A professor in the Georgia Tech School of Electrical and Computer Engineering, Howard has now launched a company, Zyrobotics, to commercialize the device, and a prototype has already been developed. The company, run by a former graduate student, won’t be the next IBM, but it will help disabled children do what all kids want to do: play video games and interact with computers.
Assistance with refining the device came through the Innovation Corps (I-Corps™), a National Science Foundation program that helps NSF-funded researchers learn about starting up a company – and by talking to potential customers, determine whether there’s really a market for what they’ve developed.
“Without I-Corps, I wouldn’t have thought to pursue this,” said Howard, who holds the title of Motorola Foundation Professor. “They showed us how to talk about the technology in terms that the general public could understand. And I-Corps made us take a step back and ask if what we had developed was really of value to potential customers.”
I-Corps Innovation
A dozen Georgia Tech teams – each composed of a faculty member, entrepreneurial lead and industry mentor – have now gone through the six-week I-Corps program. About a third of them have, like Howard, revised their plans and decided to move forward with forming a company and creating a product based on the results of NSF-supported research. The program is part of a national effort to turn research discoveries into new companies and new products, supporting economic development and building understanding of what it means to be an entrepreneur.
“Through the Innovation Corps, NSF seeks to accelerate the development of new technologies, products and processes that arise from fundamental research,” said Rathindra (Babu) DasGupta, the NSF’s program director for I-Corps. “The goals of I-Corps are to spur translation of fundamental research, to encourage collaboration between academia and industry, and to train students to understand innovation and entrepreneurship.”
The program provides mentoring and funding designed to move the results of NSF-supported research through the early stages of company formation. “NSF investments strategically strengthen the nation’s innovation ecosystem by addressing the challenges inherent in the early stages of the innovation process,” DasGupta added.
I-Corps at Georgia Tech
Because of its long experience with forming companies from university research, in July 2012 Georgia Tech was selected to be among the first institutions to become ”nodes” teaching the I-Corps curriculum. The program is basically a boot camp that shows what it’s like to form a startup company – and ensures that there’s a real market for a fledgling company’s proposed product. About 25 teams from universities around the country participate each time the program is taught at one of the I-Corps nodes, including Georgia Tech.
“The I-Corps process is very similar to the scientific method, which scientists and engineers are familiar with,” explained Keith McGreggor, who directs the I-Corps program at Georgia Tech. “We use this process to turn fiction – what you might think is true – into fact by doing experiments and testing hypotheses in the real world with customers instead of in the laboratory.”
I-Corps puts faculty members and graduate students through a pressure cooker environment that simulates a real startup. Not everyone is cut out for entrepreneurship, McGreggor noted. Faculty members often have a skill set – collaborating with other researchers, teaching students and publishing papers – that’s different from the skills needed to produce products and services that non-researchers are willing to buy.
The centerpiece of the program is “customer discovery” in which the teams must talk with at least 100 potential customers about their proposed product. This interaction with the real world almost inevitably leads to what I-Corps calls “the pivot,” which occurs when the teams, based on the customer feedback, realize they’ve been developing a product for which there isn’t a market. In many cases, that realization leads to new, and successful, directions for the technology.
“Everyone starts out with one idea about what they want to do, and they almost always change to something else that they are also capable of doing,” McGreggor said. “It can be difficult for people to switch gears, but what’s beautiful about this program is that they do switch.”
At the end of the six weeks, the teams decide whether or not to go forward with their idea. For Georgia Tech teams, fledgling companies that emerge from the process can join VentureLab, a program that helps researchers form companies, create prototypes, bring in experienced management and obtain early-stage funding. VentureLab companies can go on to be members of the Advanced Technology Development Center (ATDC), Georgia Tech’s accelerator program that helps entrepreneurs launch and build successful companies.
Marketing MOFs
Krista Walton and David Sholl used the I-Corps process to confirm the market need for metal-organic frameworks (MOFs), a new materials technology with a broad range of potential market applications. With NSF support, the researchers had developed a way to scale up the synthesis of MOFs, a class of nanomaterials, but weren’t sure what direction to take next – a classic problem for technologies that have many possible applications.
“By talking with more than 100 potential customers, we went through numerous refinements in our understanding of how we can create a sustainable business with our technology,” said Sholl, who is now chair of Georgia Tech’s School of Chemical and Biomolecular Engineering. “We saw over and over again that the issues that obsess researchers doing fundamental research and the issues that matter to customers are often not the same.”
Talking with the customers required a large investment of time, but Sholl – who is also a Georgia Research Alliance Eminent Scholar in Energy Sustainability – was pleased with the level of interest in the technology. The potential customers he and Walton interviewed also identified applications they had never considered.
As a result of the process, Sholl and Walton – an associate professor in the School of Chemical and Biomolecular Engineering – formed Inmondo Tech, and are working with several initial customers to develop a first product.
Smartphone Questions
For Gregory Abowd, the benefits of I-Corps were different. A serial entrepreneur with a record of launching successful companies, Abowd felt he knew how to commercialize technology he developed that helps connect young patients with their doctors through handheld devices. But he wanted to apply I-Corps’ systematic process to starting up a new company.
“I’ve had some successful and unsuccessful startup efforts, but I really didn’t understand what were the important elements of the successful ones,” said Abowd, who is a Regents’ and Distinguished Professor in Georgia Tech’s School of Interactive Computing. “I was intrigued with the idea of being a little more structured going into this one, because I had learned there are an infinite number of ways to make mistakes in the business world.”
The company, established as L.S.Q. LLC in Georgia, will provide a way to ask questions of smartphone users at times when they aren’t actively using their handheld devices. Building on the original purpose of the technology, which was to boost interaction with children who have chronic diseases, Abowd sees many possible applications, including surveys designed for the small screens of mobile devices.
“We’ll ask questions at a point when people are interacting with their phones, but at a point of pause,” he explained. Abowd has assembled a team and is talking with potential customers. He expects to form a joint venture with a market research firm in early 2014 and develop a product quickly.
Advice to Others
What advice do the teams give faculty members and graduate students thinking about the I-Corps opportunity?
“There is a growing network to help with commercialization, both at Georgia Tech and around the country,” noted Abowd. “A successful startup requires a lot of effort, and it’s more than a full-time job. I-Corps gives you a six-week exposure to help you determine whether this is right for you.”
I-Corps requires a large investment of time, something that can be difficult if faculty members aren’t prepared for it, Howard noted. To be successful, at least one member of the team has to be available nearly full-time during the six-week program.
“I would recommend this 100 percent, and have already talked with other faculty members about I-Corps,” she said. “This process is very different from what we normally do in research and teaching, and it has changed the way I think about what I do. It was a great experience for us.”
I-Corps teams follow a rigorous application process designed to determine whether team members are truly committed to launching and building a startup, McGreggor noted. That can be daunting.
“I-Corps simulates a startup, so it puts a lot of heat on the team to see if they are going to stay together when they get into a company,” he said. “We challenge the researchers in ways that they have probably not been challenged since they were graduate students. It is exquisitely uncomfortable for some people.”
Broader Impacts
I-Corps has also changed the way that Georgia Tech approaches startup companies. Customer discovery and early pivoting to serve the marketplace, for instance, are now at the core of Georgia Tech’s VentureLab and Flashpoint programs, which serve all researchers regardless of their funding sources, McGreggor said.
“Faculty members are forced to look into the face of a world that may not want what they have produced,” McGreggor said. “What we’ve learned is that when entrepreneurs get it wrong, it’s usually because they are building something that nobody really wants. This has really changed our approach to doing things in VentureLab.”
The I-Corps approach has also changed the role of graduate students in the startup process, and opened it more to junior faculty members. In the past, VentureLab had assumed that only tenured faculty would have the time and flexibility to commit to a startup. Now, he says, the program makes no distinction among researchers, and realizes that the graduate students involved in developing a technology may be the right team members to go forward as part of the new company. That makes creating a startup a real alternative to traditional post-graduation opportunities.
Beyond the new enterprises begun, the I-Corps program is having a larger impact on the universities whose faculty members have participated.
“Additional successes of the program have been far-reaching,” said the NSF’s DasGupta. “Faculty are taking what they learned in I-Corps about innovation and technology transfer back to their universities and training their students differently. The participation of students and post-docs in I-Corps has also had favorable impacts: they report that their employability is enhanced by their participating in I-Corps.”
The program was launched in 2011, and continues to evolve as NSF tracks the results. In addition to its teams of researchers, entrepreneurs and mentors, I-Corps is also focusing on nodes and sites to bring the concepts to a larger group of NSF researchers.
“We continue to explore ways to expand the program’s impact nationally, and at the state and local levels,” DasGupta added.
Research News
Georgia Institute of Technology
177 North Avenue
Atlanta, Georgia 30332-0181
Media Relations Contacts: John Toon (jtoon@gatech.edu) (404-894-6986) or Brett Israel (brett.israel@comm.gatech.edu) (404-385-1933).
Writer: John Toon
]]>
Research News
(404) 894-6986
]]>It’s not uncommon to see large, fast-moving robots on manufacturing floors. Humans seldom work next to them because of safety reasons. Some jobs, however, require people and robots to work together. For example, a person hanging a car door on a hinge uses a lever to guide a robot carrying the door. The power-assisting device sounds practical but isn’t easy to use.
“It turns into a constant tug of war between the person and the robot,” explains Billy Gallagher, a recent Georgia Tech Ph.D. graduate in robotics who led the project. “Both react to each other’s forces when working together. The problem is that a person’s muscle stiffness is never constant, and a robot doesn’t always know how to correctly react.”
For example, as human operators shift the lever forward or backward, the robot recognizes the command and moves appropriately. But when they want to stop the movement and hold the lever in place, people tend to stiffen and contract muscles on both sides of their arms. This creates a high level of co-contraction.
“The robot becomes confused. It doesn’t know whether the force is purely another command that should be amplified or ‘bounced’ force due to muscle co-contraction,” said Jun Ueda, Gallagher’s advisor and a professor in the Woodruff School of Mechanical Engineering. “The robot reacts regardless.”
The robot responds to that bounced force, creating vibration. The human operators also react, creating more force by stiffening their arms. The situation and vibrations become worse.
“You don’t want instability when a robot is carrying a heavy door,” said Ueda.
The Georgia Tech system eliminates the vibrations by using sensors worn on a controller’s forearm. The devices send muscle movements to a computer, which provides the robot with the operator’s level of muscle contraction. The system judges the operator's physical status and intelligently adjusts how it should interact with the human. The result is a robot that moves easily and safely.
“Instead of having the robot react to a human, we give it more information,” said Gallagher. “Modeling the operator in this way allows the robot to actively adjust to changes in the way the operator moves.”
Ueda will continue to improve the system using a $1.2 million National Robotics Initiative grant supported by a National Science Foundation grant (1317718) to better understand the mechanisms of neuromotor adaptation in human-robot physical interaction. The research is intended to benefit communities interested in the adaptive shared control approach for advanced manufacturing and process design, including automobile, aerospace and military.
“Future robots must be able to understand people better,” Ueda said. “By making robots smarter, we can make them safer and more efficient.”
The research team also includes Associate Professor Minoru Shinohara (Applied Physiology), Assistant Professor Karen Feigh (Aerospace Engineering) and Professor Emeritus Wayne Book (Mechanical Engineering).
]]>