The Robot Therapist Is Open for Business

The Robot Therbeiapist Is Open for Business The Robot Therapist Is Open for Business The Robot Therapist Is Open for BusinessProfessor-researcher-roboticist Ayanna Howard is living her dreambringing together her fascination with robotics and her childhood desire to be involved in the medical field. In so doing, she is harnessing the power of robotics to captivate and serve a very special constituency children with special needs.Collaborating with clinicians, Howard, director of Georgia Techs Human-Automation Systems Lab, and her kollektiv are developing an innovative virtual reality system that will allow various types of patient-specific therapy to take place in the home without a therapist present.The system of devices and software will replicate and replace the interactions that take place when a therapist is on site whether, for example, promoting physical movement for children with motor disabilities or behavioral responses for autistic children.The hope is that widespread deplo yment is just three or so years away, dependent on whether one of several new robotics startups develop a platform that will meet her needs something akin to downloading an app on a tablet and bring down costs from the $8,000 robot currently being used for testing to about $1,500. The child will have the robot and customized software that will include games and other interactive modules, prescribed by the clinician based on what the child is to work on.This robot, called Darwin, could provide routine direction and encouragement to people who need physical or behavioral therapy. Image Georgia TechThe first program tested was for children with cerebral palsy, who have difficulty with movement. The testing that started about 18 months ago for up to eight-week sessions took place with a therapist present. The child sits in front of a video screen to play a game called Super Pop, basically popping bubbles that move on the screen.If, for example, the child needs to work on speed of move ment, the robot, typically placed to the left of the child, will say, Hey, lets play a game. I want you to play. The robot demonstrates playing the game, all the while communicating with the child. Then the child plays, and the robot provides feedback. A follow-up could be playing the same game faster or working on anotlageher skills besides speed.One of the biggest challengesis developing the robotic system so that it functions as a human therapist in a clinical setting, providing feedback by monitoring the childs performance, says Howard.It needs to not only provide motivation like Thats a good job but also needs to recognize if the child is struggling. One typical outcome in moving faster is that another parameter gets weaker, says Howard. One example would be I might move faster but my arm isnt straight, she notes.An experienced therapist knows exactly what type of feedback to give a child to change that, she says, and thats how the system needs to operate as well.Another challe nge is creating a system thats robust but easy for non-engineers to use. The team is also addressing questions like what happens if the robot, which also includes a camera providing a visual for the clinician, is not placed in exactly the right spot and is slightly off. Maybe the robot can be trained to adapt itself to deal with an inexperienced environment, she says.Prior to widespread deployment, the system needs more games and more and larger clinical studies. A really good clinical study would be about 25 to 30 kids at the same time for something over six months, she says. The team has also just started a study for children with autism, focused on turn-taking and other activities in order to encourage appropriate behavior.Despite Howards childhood desire to go to medical school, a plan she abandoned for robotics after realizing in biology class that dissecting frogs wasnt her thing, she spent more than a decade at NASAs Jet Propulsion Laboratories, designing smarter robots using artificial intelligence for the next generation of the Mars rover. When NASA cut its research budget, she chose to join Georgia Tech to continue her research in robotics there.During the summers, she ran robotics camps for children that included both mainstream kids and children with special needs, all interacting and learning how to program robots. One year, when a teenage camper had a visual impairment, Howard began thinking about accessibility. In doing some research, she learned that an astounding percentage of children had some sort of disability.Here at camp Id been preaching that its all about diversity and inclusiveness, she says.I discovered there was this whole entire unmet need to provide children with special needs access to what we consider basic things such as education.Her first project was creating technology that would allow kids with disabilities to program robots. Then, as she came in contact with clinicians and therapists, they said its nice to get kids to progr am and open up their minds, but there were needs on the clinical side. Thats when, collaborating with clinicians, she started looking at how robotic systems could be used for therapy.Nancy Giges is an independent writer. For Further DiscussionHere Id been preaching that its all about diversity and inclusiveness, and I discovered there was this whole entire unmet need to provide children with special needs access to basic things.Prof. Ayanna Howard, Georgia Tech

Making Solar Panels More Efficient

Making Solar Panels More Efficient Making Solar Panels More Efficient A team of researchers at Massachusetts Institute of Technology has come up with a new way to capture solar energy that makes it easier to store and be used on demand at a later time.The team created a device that improves the efficiency of solar panels by using wavelengths of light that normally are wasted because they cannot be captured by conventional photovoltaic cells. In this new system, the sun heats a high-temperature material, a two-layer absorber-emitter device placed over the PV cells. The outer sunlight-facing layer, the absorber, includes an array of multi-walled carbon nanotubes that efficiently absorbs the lights energy and turns it into heat. A bonded layer of silicon/silicon dioxide photonic crystals, the emitter, is engineered to convert the heat back into light that can then be captured by the PV cells. This allows much more of the energy in the sunlight to be turned into electricity.This new syst em combines the advantages of solar photovoltaic systems, which turn sunlight directly into electricity, and solar thermal systems, beneficial for delayed use because heat is more easily stored than electricity. The basic concept has been explored for several years, according to the team.Earlier StudiesA lot of work has been done on the theoretical design of surfaces for solar thermophotovoltaic systems (STPVs) and fabrication of single components for potential integration in these systems, says team member Andrej Lenert, an MIT graduate student who expects to be awarded his PhD in mechanical engineering this spring.Schematic of the planar STPV layout. Incoming solar radiation is converted to heat at the absorber heat is selectively radiated by the emitter, and converted to electrical power at the PV cell.Image MIT.eduLenert has been involved with STPV efforts at MIT ever since the university opened the Solid-State Solar Thermal Energy Conversion (S3TEC) Center in 2010, but his inte rest goes back even further to a radiation class. I was drawn to this work anfangsbuchstabely because of the elegance of the concept and later because of the multi-disciplinary nature of its practical implementation, he says. My interest in renewable power generation stems as far back as my interest in pursuing an engineering degree. He expects to continue research in this area after graduation.While the earlier studies have suggested efficiencies as high as 40%, experiments remained below 1%, Lenert says. The large discrepancy is in parte due to the challenging experimental nature of spectral engineering at high temperatures. It is also in part due to fact that the overall system efficiency is highly dependent on the performance of each one of the energy conversion steps and components, just like in a conventional solar cell, except with the added spectral conversion steps in the hot absorber-emitter.He says the team came up with the idea for the absorber-emitter after developing a framework to identify which parts of the spectrum are most critical to the success of an STPV system. We then tuned the spectral properties of the absorber-emitter using carbon nanotubes and silicon/silicon dioxide photonic crystals to target these properties and achieve the improved performance, he says.Key to the breakthrough was an understanding of the interplay between the use of structure at small scales to tune spectral properties and macroscale device design.Lenerts team has produced an initial test device with a measured efficiency of 3.2%, and they say with further work they expect to be able to reach 20% efficiency, enough for a commercially viable product.Further OptimizationIn their experiments using simulated sunlight, the researchers found peak efficiency came when the intensity was equivalent to a focusing system that concentrates sunlight by a factor of 750. This level of concentration is already much lower than in previous attempts at STPV systems, which concentrat ed sunlight by a factor of several thousand. But the MIT researchers say that after further optimization, it should be possible to get the same kind of enhancement at even lower sunlight concentrations, making the systems easier to operate.Lenert says this is because the research center is currently working on getting even better control of the thermally-driven spectral conversion process using wavelength and angular selective surfaces. This selectivity will lower the required level of solar concentration in two ways Control over re-emission losses from the absorber and a more efficient TPV process that will contribute to lowering the input solar power needed to reach the same operating temperature.If the team achieves its goal of generating power from sunlight both efficiently and on demand from an STPV system, it could have a major impact on the way society uses solar power or at least provide another renewable option for applications when solar thermal plants or photovoltaics can not meet the requirements, Lenert says.Nancy S. Giges is an independent writer.For Further Discussion The large discrepancy is in part due to the challenging experimental nature of spectral engineering at high temperatures.Andrej Lenert, MIT graduate student

Video Edith Stern, 2012 ASME Kate Gleason Award Recipient

Video Edith Stern, 2012 ASME Kate Gleason Award Recipient Video Edith Stern, 2012 ASME Kate Gleason Award Recipient Video Edith Stern, 2012 Kate Gleason AwardEdith Stern climbed the IBM ladder from summer intern to distinguished engineer. She was a child prodigy in mathematics, and developed a passion for number theory. At IBM, she has solved challenging problems in the telephony of the 1970s through todays multimedia digital conferencing. Proof of her proficiency as an inventor comes in the form of 110 patents.Ms. Sterns contributions in applied mathematics span many industries. She has helped put tablet computers on 18-wheeler trucks, developed a commercial playback ordnungsprinzip for a television network, and initiated projects using radio frequency identification for health care companies.Ms. Stern received a masters degree in mathematics at Michigan State University in East Lansing. She received her bachelors degree in mathematics at Florida Atlantic University in Boca Raton.Re ad more about the 2012 Recipients of ASME Honors and Awards The copyright of this program is owned by ASME.