"viernes, 23 de abril de 2010
Personal Mobility Robot Operated by Wii Controller
Personal Mobility Robot Operated by Wii Controller: "Japanese researchers demonstrate a robotic wheelchair operated with Wii game controller"
"
Boston Dynamics Biped Robot Petman Achieves 4.4 mph
Boston Dynamics Biped Robot Petman Achieves 4.4 mph: "Boston Dynamics has released a new video showing its Petman biped robot achieving 4.4 mph on a treadmill. So now the question is, can it jog?
"
"jueves, 22 de abril de 2010
México brilla en el Mundial de futbol... de robots - Deportes - CNNMéxico.com
México brilla en el Mundial de futbol... de robots - Deportes - CNNMéxico.com
ATLANTA (CNN) — A pocos meses de la Copa del Mundo de futbol, México tuvo su primer ensayo mundial y podría decirse que el equipo funcionó como una máquina. Literalmente, porque fue en una competencia internacional de robots en la que los representantes mexicanos brillaron.
El torneo de Breakaway, una curiosa mezcla de futbol soccer y Nascar, fue parte de una competencia mundial de robótica celebrada en Atlanta la semana pasada y organizada por FIRST, una organización sin ánimo de lucro que promueve la ciencia y tecnología entre los jóvenes.
Participaron 530 robots construidos por más de 10,000 estudiantes de 30 países, entre ellos alumnos de la preparatoria de la Universidad Panamericana (UP) y del Tecnológico de Monterrey campus Toluca.
Panteras, el equipo de la UP, participa en estas competencias desde hace cuatro años y ha ganado varios reconocimientos, como el premio al Novato del Año en 2007.
“Queremos cambiar el paradigma de México, queremos que más gente se involucre y pueda participar”, dice Andrés Carrasco, uno de los integrantes de Panteras encargado de buscar financiación para el proyecto.
Solamente inscribirse cuesta 6,000 dólares, y en la preparación y construcción del robot pueden llegar a invertirse unos 25,000 dólares. “Nosotros tenemos que conseguir los materiales. Las empresas nos pueden ayudar con dinero y en especie: GM nos da los motores, por ejemplo, y 3M los materiales desechables”, explica Sebastián Montes, otro de los miembros del equipo.
Para construir estos robots se necesitan conocimientos de mecánica, electrónica y programación, aunque no hace falta ser un genio para participar. “Lo que se necesitan son ganas”, señala Sebastián. “Hay que meterse, practicar, adquirir experiencia y aprender”. Además, los estudiantes cuentan con la ayuda de los mentores, profesionales de empresas que colaboran con los alumnos.
El otro equipo mexicano que compitió fue el Tecbot, de la preparatoria del TEC sede Toluca y que debutó este año con bombos y platillos. Su delegación, con más de 40 estudiantes y 14 mentores, fue una de las más numerosas... y también "ruidosas", con sus llamativas indumentarias y cánticos constantes.
“Es una experiencia internacional muy buena. Me interesó porque mezcla la robótica con los negocios”, dice Samantha Carmona, una de las encargadas de la mercadotecnia y publicidad de Tecbot. Carmona comparó el funcionamiento del equipo con una microempresa en la que al final “nos volvimos como una familia”.
Los miembros de Tecbot quedaron tan entusiasmados con el proyecto que ya planean organizar un campeonato regional de Latinoamérica.
Aunque los equipos mexicanos no clasificaron para las finales de Breakaway, en la ceremonia de entrega de premios recibieron varios reconocimientos. Panteras ganó el premio Team Spirit, por su “entusiasmo y espíritu a través de un compañerismo excepcional y el trabajo en equipo”, mientras que Tecbot se llevó el Rookie Inspiration, por su “destacado esfuerzo en el trabajo en la comunidad y en el reclutamiento de estudiantes”.
Al recibir los premios, las delegaciones mexicanas lo festejaron con una gran algarabía que contagió a las gradas. Y es que México demostró que, aunque no siempre se gane, su afición es una de las mejores del mundo... también en robots.
lunes, 19 de abril de 2010
Robotic therapy helps stroke patients regain function
Robotic therapy helps stroke patients regain function: "Stroke patients who received robot-assisted therapy were able to regain some ability to use their arms, even if the stroke had occurred years earlier, according to a study published April 16 in the online issue of The New England Journal of Medicine.
The study, which examined the effectiveness of a class of robotic devices developed at MIT, found that in chronic stroke survivors, robot-assisted therapy led to modest improvements in upper-body motor function and quality of life six months after active therapy was completed; these improvements were significant when compared with a group of stroke patients who received the traditional treatment. Moreover, the robotic therapy — which involves a more intense regimen of activity than traditional stroke therapy — did not increase total health-care costs per stroke patient, and could make intensive therapy available to more people, say the researchers who led the study.
The study results also challenge the notion that physical therapy only benefits stroke patients within the first six months after the stroke, says Albert Lo, a neurologist at the Providence VA Medical Center who led the study.
“There are nearly six million stroke patients in the U.S. with chronic deficits,” says Lo. “We’ve shown that with the right therapy, they can see improvements in movement, everyday function and quality of life.”
Mind and body
The study, conducted at four Veterans Affairs (VA) hospitals, found that patients who used the MIT robotic devices for 12 weeks experienced a small but significant gain in arm function. Another group of patients who received high-intensity therapy from a therapist, which matched the number and intensity of the robot movements, showed similar improvements.
Hermano Igo Krebs, a principal research scientist in MIT’s Department of Mechanical Engineering who developed the MIT-Manus robot, has been working on robotic therapy since his graduate student years at MIT almost 20 years ago. In his early studies, he and his colleague, Professor Neville Hogan, found that it’s important for stroke patients to make a conscious effort during physical therapy.
The MIT-Manus system, which Krebs started developing more than 20 years ago, is based on that principle. The patient grasps a robotic joystick that guides the patient’s arm, wrist or hand as he or she tries to make specific movements, helping the brain form new connections that will eventually help the patient relearn to move the limb on his or her own.
In the New England Journal of Medicine study, researchers at VA hospitals in Baltimore, Seattle, West Haven, Conn., and Gainesville, Fla., compared the MIT-Manus system to a high-intensity rehab program delivered by a human therapist, which was designed specifically for this study.
Each group included about 50 patients, who were also compared with a group of 28 stroke patients who received so-called “usual care” — general health care and three hours per week of traditional physical therapy for their stroke-damaged limb.
Patients using the MIT-Manus system grasp a joystick-like handle connected to a computer monitor that displays tasks similar to those in simple video games. In a typical task, the subject attempts to move the robot handle toward a moving or stationary target shown on the computer monitor. If the person starts moving in the wrong direction or does not move, the robotic arm gently nudges his or her arm in the right direction.
“The ability to be interactive is critical,” says Krebs. “We program the robot to only give assistance as needed.”
Patients in the study received therapy three times a week for 12 weeks, and during each hour-long session, they made hundreds of repetitive motions with their arms. At the end of 12 weeks, tests revealed a small but statistically significant improvement in quality of life, and a modest improvement in arm function. When the subjects were tested again at 36 weeks, both the robot therapy group and intensive human-assisted therapy group showed improvement in arm movement and strength, everyday function and quality of life compared to the usual-care group.
The high-intensity, interactive physical therapy offered to patients who did not receive robot-assisted therapy was developed specifically for comparison purposes for this study, and is not generally available. Furthermore, the physical demands on the therapist make it unlikely that it will ever be widely used.
“If you can get a therapist to work at that pace with a patient, certainly the benefits are roughly the same, and we showed this benefit when we designed this intensive comparison group, but it’s not practical,” says Krebs. “Robotics and automation technology are ideal for this kind of highly repetitive tasks. We’re using robotic technology to create a tool for the therapist to afford this kind of high-intensity therapy while maintaining the therapist supervisory role, deciding what is right for a particular patient.”
This particular study was designed to test the effects of only conventional therapy versus only robotic therapy, but Bruce Dobkin, a neurologist at the UCLA Stroke Center, says the best approach may end up being a combination of those two strategies. “If robotic therapy is going to be helpful, you need to find a more integrated way to use the robotic device,” he says.
The value of robots
Another way to make robotic therapy more practical could be to lower the costs, says Dobkin, who was part of the data-safety monitoring committee that supervised the research. In the VA study, the robotic therapy cost an average of $9,977 per patient, and the intensive nonrobotic therapy cost $8,269 per patient. However, overall healthcare per-patient costs, including costs for those who received only usual care, were not very different over the total 36-week study period — $15,562 per patient for robot-assisted therapy, $15,605 for intensive nonrobotic therapy, and $14,343 for usual care.
Krebs believes that once the robotic devices can be mass-produced, which he expects will occur within the next 10 years, the costs will drop. “What you have to do is make more of them, and that will drive down costs to a point where people can have them in their homes,” he says.
Krebs is also encouraged by the fact that many of the patients in the study had either suffered multiple strokes or had experienced their strokes many years earlier, yet still showed improvement. “We put the bar very high,” he says. “If we worked with patients sooner after their first stroke, we may get even better results.” He is now working with doctors to plan such a study.
Krebs and his collaborators are also studying whether the MIT-Manus could help patients with cerebral palsy, multiple sclerosis and spinal cord injury.
"
The study, which examined the effectiveness of a class of robotic devices developed at MIT, found that in chronic stroke survivors, robot-assisted therapy led to modest improvements in upper-body motor function and quality of life six months after active therapy was completed; these improvements were significant when compared with a group of stroke patients who received the traditional treatment. Moreover, the robotic therapy — which involves a more intense regimen of activity than traditional stroke therapy — did not increase total health-care costs per stroke patient, and could make intensive therapy available to more people, say the researchers who led the study.
The study results also challenge the notion that physical therapy only benefits stroke patients within the first six months after the stroke, says Albert Lo, a neurologist at the Providence VA Medical Center who led the study.
“There are nearly six million stroke patients in the U.S. with chronic deficits,” says Lo. “We’ve shown that with the right therapy, they can see improvements in movement, everyday function and quality of life.”
Mind and body
The study, conducted at four Veterans Affairs (VA) hospitals, found that patients who used the MIT robotic devices for 12 weeks experienced a small but significant gain in arm function. Another group of patients who received high-intensity therapy from a therapist, which matched the number and intensity of the robot movements, showed similar improvements.
Hermano Igo Krebs, a principal research scientist in MIT’s Department of Mechanical Engineering who developed the MIT-Manus robot, has been working on robotic therapy since his graduate student years at MIT almost 20 years ago. In his early studies, he and his colleague, Professor Neville Hogan, found that it’s important for stroke patients to make a conscious effort during physical therapy.
The MIT-Manus system, which Krebs started developing more than 20 years ago, is based on that principle. The patient grasps a robotic joystick that guides the patient’s arm, wrist or hand as he or she tries to make specific movements, helping the brain form new connections that will eventually help the patient relearn to move the limb on his or her own.
In the New England Journal of Medicine study, researchers at VA hospitals in Baltimore, Seattle, West Haven, Conn., and Gainesville, Fla., compared the MIT-Manus system to a high-intensity rehab program delivered by a human therapist, which was designed specifically for this study.
Each group included about 50 patients, who were also compared with a group of 28 stroke patients who received so-called “usual care” — general health care and three hours per week of traditional physical therapy for their stroke-damaged limb.
Patients using the MIT-Manus system grasp a joystick-like handle connected to a computer monitor that displays tasks similar to those in simple video games. In a typical task, the subject attempts to move the robot handle toward a moving or stationary target shown on the computer monitor. If the person starts moving in the wrong direction or does not move, the robotic arm gently nudges his or her arm in the right direction.
“The ability to be interactive is critical,” says Krebs. “We program the robot to only give assistance as needed.”
Patients in the study received therapy three times a week for 12 weeks, and during each hour-long session, they made hundreds of repetitive motions with their arms. At the end of 12 weeks, tests revealed a small but statistically significant improvement in quality of life, and a modest improvement in arm function. When the subjects were tested again at 36 weeks, both the robot therapy group and intensive human-assisted therapy group showed improvement in arm movement and strength, everyday function and quality of life compared to the usual-care group.
The high-intensity, interactive physical therapy offered to patients who did not receive robot-assisted therapy was developed specifically for comparison purposes for this study, and is not generally available. Furthermore, the physical demands on the therapist make it unlikely that it will ever be widely used.
“If you can get a therapist to work at that pace with a patient, certainly the benefits are roughly the same, and we showed this benefit when we designed this intensive comparison group, but it’s not practical,” says Krebs. “Robotics and automation technology are ideal for this kind of highly repetitive tasks. We’re using robotic technology to create a tool for the therapist to afford this kind of high-intensity therapy while maintaining the therapist supervisory role, deciding what is right for a particular patient.”
This particular study was designed to test the effects of only conventional therapy versus only robotic therapy, but Bruce Dobkin, a neurologist at the UCLA Stroke Center, says the best approach may end up being a combination of those two strategies. “If robotic therapy is going to be helpful, you need to find a more integrated way to use the robotic device,” he says.
The value of robots
Another way to make robotic therapy more practical could be to lower the costs, says Dobkin, who was part of the data-safety monitoring committee that supervised the research. In the VA study, the robotic therapy cost an average of $9,977 per patient, and the intensive nonrobotic therapy cost $8,269 per patient. However, overall healthcare per-patient costs, including costs for those who received only usual care, were not very different over the total 36-week study period — $15,562 per patient for robot-assisted therapy, $15,605 for intensive nonrobotic therapy, and $14,343 for usual care.
Krebs believes that once the robotic devices can be mass-produced, which he expects will occur within the next 10 years, the costs will drop. “What you have to do is make more of them, and that will drive down costs to a point where people can have them in their homes,” he says.
Krebs is also encouraged by the fact that many of the patients in the study had either suffered multiple strokes or had experienced their strokes many years earlier, yet still showed improvement. “We put the bar very high,” he says. “If we worked with patients sooner after their first stroke, we may get even better results.” He is now working with doctors to plan such a study.
Krebs and his collaborators are also studying whether the MIT-Manus could help patients with cerebral palsy, multiple sclerosis and spinal cord injury.
"
viernes, 16 de abril de 2010
A Bayesian Exploration-Exploitation Approach for Optimal Online Sensing and Planning with a Visually Guided Mobile Robot
A Bayesian Exploration-Exploitation Approach for Optimal Online Sensing and Planning with a Visually Guided Mobile Robot from Ruben Martinez-Cantin on Vimeo.
We address the problem of online path planning for optimal sensing with a mobile robot. The objective of the robot is to learn the most about its pose and the environment given time constraints. We use a POMDP with a utility function that depends on the belief state to model the finite horizon planning problem.We replan as the robot progresses throughout the environment. The POMDP is highdimensional, continuous, non-differentiable, nonlinear, non- Gaussian and must be solved in real-time. Most existing techniques for stochastic planning and reinforcement learning are therefore inapplicable. To solve this extremely complex problem, we propose a Bayesian optimization method that dynamically trades off exploration (minimizing uncertainty in unknown parts of the policy space) and exploitation (capitalizing on the current best solution). We demonstrate our approach with a visually-guide mobile robot. The solution proposed here is also applicable to other closelyrelated domains, including active vision, sequential experimental design, dynamic sensing and calibration with mobile sensors.
Visual Map-less Navigation based on Homographies
Visual Map-less Navigation based on Homographies: "
Comparison between two different turns (second and tenth) of the square with visual correction. The errors in the trajectory are so small that they can not be apreciated in the movie. The image seems to be blurred because the two videos are superimposed. Check users.isr.ist.utl.pt/~rmcantin/pmwiki/pmwiki.php/Research/Mapless for details.
Cast: Ruben Martinez-Cantin
"Visual Map-less Navigation based on Homographies from Ruben Martinez-Cantin on Vimeo.
Suscribirse a:
Entradas (Atom)