More stories

  • in

    The robot smiled back

    While our facial expressions play a huge role in building trust, most robots still sport the blank and static visage of a professional poker player. With the increasing use of robots in locations where robots and humans need to work closely together, from nursing homes to warehouses and factories, the need for a more responsive, facially realistic robot is growing more urgent.
    Long interested in the interactions between robots and humans, researchers in the Creative Machines Lab at Columbia Engineering have been working for five years to create EVA, a new autonomous robot with a soft and expressive face that responds to match the expressions of nearby humans. The research will be presented at the ICRA conference on May 30, 2021, and the robot blueprints are open-sourced on Hardware-X (April 2021).
    “The idea for EVA took shape a few years ago, when my students and I began to notice that the robots in our lab were staring back at us through plastic, googly eyes,” said Hod Lipson, James and Sally Scapa Professor of Innovation (Mechanical Engineering) and director of the Creative Machines Lab.
    Lipson observed a similar trend in the grocery store, where he encountered restocking robots wearing name badges, and in one case, decked out in a cozy, hand-knit cap. “People seemed to be humanizing their robotic colleagues by giving them eyes, an identity, or a name,” he said. “This made us wonder, if eyes and clothing work, why not make a robot that has a super-expressive and responsive human face?”
    While this sounds simple, creating a convincing robotic face has been a formidable challenge for roboticists. For decades, robotic body parts have been made of metal or hard plastic, materials that were too stiff to flow and move the way human tissue does. Robotic hardware has been similarly crude and difficult to work with — circuits, sensors, and motors are heavy, power-intensive, and bulky.
    The first phase of the project began in Lipson’s lab several years ago when undergraduate student Zanwar Faraj led a team of students in building the robot’s physical “machinery.” They constructed EVA as a disembodied bust that bears a strong resemblance to the silent but facially animated performers of the Blue Man Group. EVA can express the six basic emotions of anger, disgust, fear, joy, sadness, and surprise, as well as an array of more nuanced emotions, by using artificial “muscles” (i.e. cables and motors) that pull on specific points on EVA’s face, mimicking the movements of the more than 42 tiny muscles attached at various points to the skin and bones of human faces. More

  • in

    Artificial neurons recognize biosignals in real time

    Current neural network algorithms produce impressive results that help solve an incredible number of problems. However, the electronic devices used to run these algorithms still require too much processing power. These artificial intelligence (AI) systems simply cannot compete with an actual brain when it comes to processing sensory information or interactions with the environment in real time.
    Neuromorphic chip detects high-frequency oscillations
    Neuromorphic engineering is a promising new approach that bridges the gap between artificial and natural intelligence. An interdisciplinary research team at the University of Zurich, the ETH Zurich, and the UniversityHospital Zurich has used this approach to develop a chip based on neuromorphic technology that reliably and accurately recognizes complex biosignals. The scientists were able to use this technology to successfully detect previously recorded high-frequency oscillations (HFOs). These specific waves, measured using an intracranial electroencephalogram (iEEG), have proven to be promising biomarkers for identifying the brain tissue that causes epileptic seizures.
    Complex, compact and energy efficient
    The researchers first designed an algorithm that detects HFOs by simulating the brain’s natural neural network: a tiny so-called spiking neural network (SNN). The second step involved imple-menting the SNN in a fingernail-sized piece of hardware that receives neural signals by means of electrodes and which, unlike conventional computers, is massively energy efficient. This makes calculations with a very high temporal resolution possible, without relying on the internet or cloud computing. “Our design allows us to recognize spatiotemporal patterns in biological signals in real time,” says Giacomo Indiveri, professor at the Institute for Neuroinformatics of UZH and ETH Zur-ich.
    Measuring HFOs in operating theaters and outside of hospitals
    The researchers are now planning to use their findings to create an electronic system that reliably recognizes and monitors HFOs in real time. When used as an additional diagnostic tool in operating theaters, the system could improve the outcome of neurosurgical interventions.
    However, this is not the only field where HFO recognition can play an important role. The team’s long-term target is to develop a device for monitoring epilepsy that could be used outside of the hospital and that would make it possible to analyze signals from a large number of electrodes over several weeks or months. “We want to integrate low-energy, wireless data communications in the design — to connect it to a cellphone, for example,” says Indiveri. Johannes Sarnthein, a neurophysiologist at UniversityHospital Zurich, elaborates: “A portable or implantable chip such as this could identify periods with a higher or lower rate of incidence of seizures, which would enable us to deliver personalized medicine.” This research on epilepsy is being conducted at the Zurich Center of Epileptology and Epilepsy Surgery, which is run as part of a partnership between UniversityHospital Zurich, the Swiss Epilepsy Clinic and the University Children’s Hospital Zurich.
    Story Source:
    Materials provided by University of Zurich. Note: Content may be edited for style and length. More

  • in

    Mathematical model developed to prevent botulism

    For years, food producers who make lightly preserved, ready-to-eat food have had to follow a set of guidelines to stop growth of Clostridium botulinum bacteria and production of a strong neurotoxin. The toxin can cause a serious illness called botulism.
    For refrigerated products, the guidelines for controlling Clostridium botulinum indicate that the water contained in the products should have a salt content of at least 3.5%. Unfortunately, this hampers efforts to develop salt-reduced products, even though such products would benefit public health, as most consumers eat more salt than recommended.
    If food producers want to launch products that contain e.g. less salt, they have had to conduct laboratory experiments to document that such a change in recipe will not compromise food safety. This is a time-consuming and costly process.
    Reduced need for costly product testing
    Researchers at the National Food Institute have now developed a mathematical model, which replaces costly laboratory experiments. The industry has been asking for this model for years. The new model can predict whether a particular recipe for chilled products can prevent the growth of Clostridium botulinum and production of the toxin.
    The model is the most comprehensive of its kind in the world and can show how storage temperature, pH, salt and the use of five different preservatives (such as acetic and lactic acids) affect potential bacterial growth and production of the toxin. Previous models have only incorporated the effect of half of these factors. More

  • in

    Spacetime crystals proposed by placing space and time on an equal footing

    A Penn State scientist studying crystal structures has developed a new mathematical formula that may solve a decades-old problem in understanding spacetime, the fabric of the universe proposed in Einstein’s theories of relativity.
    “Relativity tells us space and time can mix to form a single entity called spacetime, which is four-dimensional: three space-axes and one time-axis,” said Venkatraman Gopalan, professor of materials science and engineering and physics at Penn State. “However, something about the time-axis sticks out like sore thumb.”
    For calculations to work within relativity, scientists must insert a negative sign on time values that they do not have to place on space values. Physicists have learned to work with the negative values, but it means that spacetime cannot be dealt with using traditional Euclidean geometry and instead must be viewed with the more complex hyperbolic geometry.
    Gopalan developed a two-step mathematical approach that allows the differences between space and time to be blurred, removing the negative sign problem, serving as a bridge between the two geometries.
    “For more than 100 years, there has been an effort to put space and time on the same footing,” Gopalan said. “But that has really not happened because of this minus sign. This research removes that problem at least in special relativity. Space and time are truly on the same footing in this work.” The paper, published today (May 27) in the journal Acta Crystallographica A, is accompanied by a commentary in which two physicists write that Gopalan’s approach may hold the key to unifying quantum mechanics and gravity, two foundational fields of physics that are yet to be fully unified.
    “Gopalan’s idea of general relativistic spacetime crystals and how to obtain them is both powerful and broad,” said Martin Bojowald, professor of physics at Penn State. “This research, in part, presents a new approach to a problem in physics that has remained unresolved for decades.”
    In addition to providing a new approach to relate spacetime to traditional geometry, the research has implications for developing new structures with exotic properties, known as spacetime crystals.
    Crystals contain repeating arrangement of atoms, and in recent years scientists have explored the concept of time crystals, in which the state of a material changes and repeats in time as well, like a dance. However, time is disconnected from space in those formulations. The method developed by Gopalan would allow for a new class of spacetime crystals to be explored, where space and time can mix.
    “These possibilities could usher in an entirely new class of metamaterials with exotic properties otherwise not available in nature, besides understanding the fundamental attributes of a number of dynamical systems,” said Avadh Saxena, a physicist at Los Alamos National Laboratory.
    Gopalan’s method involves blending two separate observations of the same event. Blending occurs when two observers exchange time coordinates but keep their own space coordinates. With an additional mathematical step called renormalization, this leads to “renormalized blended spacetime.”
    “Let’s say I am on the ground and you are flying on the space station, and we both observe an event like a comet fly by,” Gopalan said. “You make your measurement of when and where you saw it, and I make mine of the same event, and then we compare notes. I then adopt your time measurement as my own, but I retain my original space measurement of the comet. You in turn adopt my time measurement as your own, but retain your own space measurement of the comet. From a mathematical point of view, if we do this blending of our measurements, the annoying minus sign goes away.”
    The National Science Foundation funded this research.
    Story Source:
    Materials provided by Penn State. Note: Content may be edited for style and length. More

  • in

    Improving computer vision for AI

    Researchers from UTSA, the University of Central Florida (UCF), the Air Force Research Laboratory (AFRL) and SRI International have developed a new method that improves how artificial intelligence learns to see.
    Led by Sumit Jha, professor in the Department of Computer Science at UTSA, the team has changed the conventional approach employed in explaining machine learning decisions that relies on a single injection of noise into the input layer of a neural network.
    The team shows that adding noise — also known as pixilation — along multiple layers of a network provides a more robust representation of an image that’s recognized by the AI and creates more robust explanations for AI decisions. This work aids in the development of what’s been called “explainable AI” which seeks to enable high-assurance applications of AI such as medical imaging and autonomous driving.
    “It’s about injecting noise into every layer,” Jha said. “The network is now forced to learn a more robust representation of the input in all of its internal layers. If every layer experiences more perturbations in every training, then the image representation will be more robust and you won’t see the AI fail just because you change a few pixels of the input image.”
    Computer vision — the ability to recognize images — has many business applications. Computer vision can better identify areas of concern in the livers and brains of cancer patients. This type of machine learning can also be employed in many other industries. Manufacturers can use it to detect defection rates, drones can use it to help detect pipeline leaks, and agriculturists have begun using it to spot early signs of crop disease to improve their yields.
    Through deep learning, a computer is trained to perform behaviors, such as recognizing speech, identifying images or making predictions. Instead of organizing data to run through set equations, deep learning works within basic parameters about a data set and trains the computer to learn on its own by recognizing patterns using many layers of processing.
    The team’s work, led by Jha, is a major advancement to previous work he’s conducted in this field. In a 2019 paper presented at the AI Safety workshop co-located with that year’s International Joint Conference on Artificial Intelligence (IJCAI), Jha, his students and colleagues from the Oak Ridge National Laboratory demonstrated how poor conditions in nature can lead to dangerous neural network performance. A computer vision system was asked to recognize a minivan on a road, and did so correctly. His team then added a small amount of fog and posed the same query again to the network: the AI identified the minivan as a fountain. As a result, their paper was a best paper candidate.
    In most models that rely on neural ordinary differential equations (ODEs), a machine is trained with one input through one network, and then spreads through the hidden layers to create one response in the output layer. This team of UTSA, UCF, AFRL and SRI researchers use a more dynamic approach known as a stochastic differential equations (SDEs). Exploiting the connection between dynamical systems to show that neural SDEs lead to less noisy, visually sharper, and quantitatively robust attributions than those computed using neural ODEs.
    The SDE approach learns not just from one image but from a set of nearby images due to the injection of the noise in multiple layers of the neural network. As more noise is injected, the machine will learn evolving approaches and find better ways to make explanations or attributions simply because the model created at the onset is based on evolving characteristics and/or the conditions of the image. It’s an improvement on several other attribution approaches including saliency maps and integrated gradients.
    Jha’s new research is described in the paper “On Smoother Attributions using Neural Stochastic Differential Equations.” Fellow contributors to this novel approach include UCF’s Richard Ewetz, AFRL’s Alvaro Velazquez and SRI’s Sumit Jha. The lab is funded by the Defense Advanced Research Projects Agency, the Office of Naval Research and the National Science Foundation. Their research will be presented at the 2021 IJCAI, a conference with about a 14% acceptance rate for submissions. Past presenters at this highly selective conference have included Facebook and Google.
    “I am delighted to share the fantastic news that our paper on explainable AI has just been accepted at IJCAI,” Jha added. “This is a big opportunity for UTSA to be part of the global conversation on how a machine sees.”
    Story Source:
    Materials provided by University of Texas at San Antonio. Original written by Milady Nazir. Note: Content may be edited for style and length. More

  • in

    The path to more human-like robot object manipulation skills

    What if a robot could organize your closet or chop your vegetables? A sous chef in every home could someday be a reality.
    Still, while advances in artificial intelligence and machine learning have made better robotics possible, there is still quite a wide gap between what humans and robots can do. Closing that gap will require overcoming a number of obstacles in robot manipulation, or the ability of robots to manipulate environments and adapt to changing stimuli.
    Ph.D. candidate Jinda Cui and Jeff Trinkle, Professor and Chair of the Department of Computer Science and Engineering at Lehigh University, are interested in those challenges. They work in an area called learned robot manipulation, in which robots are “trained” through machine learning to manipulate objects and environments like humans do.
    “I’ve always felt that for robots to be really useful they have to pick stuff up, they have to be able to manipulate it and put things together and fix things, to help you off the floor and all that,” says Trinkle who has conducted decades of research in robot manipulation and is well known for his pioneering work in simulating multibody systems under contact constraints. “It takes so many technical areas together to look at a problem like that.”
    “In robot manipulation, learning is a promising alternative to traditional engineering methods and has demonstrated great success, especially in pick-and-place tasks,” says Cui, whose work has been focused on the intersection of robot manipulation and machine learning. “Although many research questions still need to be answered, learned robot manipulation could potentially bring robot manipulators into our homes and businesses. Maybe we will see robots mopping our tables or organizing closets in the near future.”
    In a review article in Science Robotics called “Toward next-generation learned robot manipulation,” Cui and Trinkle summarize, compare and contrast research in learned robot manipulation through the lens of adaptability and outline promising research directions for the future. More

  • in

    Slender robotic finger senses buried items

    Over the years, robots have gotten quite good at identifying objects — as long as they’re out in the open.
    Discerning buried items in granular material like sand is a taller order. To do that, a robot would need fingers that were slender enough to penetrate the sand, mobile enough to wriggle free when sand grains jam, and sensitive enough to feel the detailed shape of the buried object.
    MIT researchers have now designed a sharp-tipped robot finger equipped with tactile sensing to meet the challenge of identifying buried objects. In experiments, the aptly named Digger Finger was able to dig through granular media such as sand and rice, and it correctly sensed the shapes of submerged items it encountered. The researchers say the robot might one day perform various subterranean duties, such as finding buried cables or disarming buried bombs.
    The research will be presented at the next International Symposium on Experimental Robotics. The study’s lead author is Radhen Patel, a postdoc in MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL). Co-authors include CSAIL PhD student Branden Romero, Harvard University PhD student Nancy Ouyang, and Edward Adelson, the John and Dorothy Wilson Professor of Vision Science in CSAIL and the Department of Brain and Cognitive Sciences.
    Seeking to identify objects buried in granular material — sand, gravel, and other types of loosely packed particles — isn’t a brand new quest. Previously, researchers have used technologies that sense the subterranean from above, such as Ground Penetrating Radar or ultrasonic vibrations. But these techniques provide only a hazy view of submerged objects. They might struggle to differentiate rock from bone, for example.
    “So, the idea is to make a finger that has a good sense of touch and can distinguish between the various things it’s feeling,” says Adelson. “That would be helpful if you’re trying to find and disable buried bombs, for example.” Making that idea a reality meant clearing a number of hurdles. More

  • in

    The world's smallest fruit picker controlled by artificial intelligence

    The goal of Kaare Hartvig Jensen, Associate Professor at DTU Physics, was to reduce the need for harvesting, transporting, and processing crops for the production of biofuels, pharmaceuticals, and other products. The new method of extracting the necessary substances, which are called plant metabolites, also eliminates the need for chemical and mechanical processes.
    Plant metabolites consist of a wide range of extremely important chemicals. Many, such as the malaria drug artemisinin, have remarkable therapeutic properties, while others, like natural rubber or biofuel from tree sap, have mechanical properties.
    Harvesting cell by cell
    Because most plant metabolites are isolated in individual cells, the method of extracting the metabolites is also important, since the procedure affects both product purity and yield.
    Usually the extraction involves grinding, centrifugation, and chemical treatment using solvents. This results in considerable pollution, which contributes to the high financial and environmental processing costs.
    “All the substances are produced and stored inside individual cells in the plant. That’s where you have to go in if you want the pure material. When you harvest the whole plant or separate the fruit from the branches, you also harvest a whole lot of tissue that doesn’t contain the substance you’re interested in,” explains Kaare Hartvig Jensen. More