Posted on

Will artificial intelligence become conscious?

(Credit: EPFL/Blue Brain Project)

By Subhash Kak, Regents Professor of Electrical and Computer Engineering, Oklahoma State University

Forget about today’s modest incremental advances in artificial intelligence, such as the increasing abilities of cars to drive themselves. Waiting in the wings might be a groundbreaking development: a machine that is aware of itself and its surroundings, and that could take in and process massive amounts of data in real time. It could be sent on dangerous missions, into space or combat. In addition to driving people around, it might be able to cook, clean, do laundry — and even keep humans company when other people aren’t nearby.

A particularly advanced set of machines could replace humans at literally all jobs. That would save humanity from workaday drudgery, but it would also shake many societal foundations. A life of no work and only play may turn out to be a dystopia.

Conscious machines would also raise troubling legal and ethical problems. Would a conscious machine be a “person” under law and be liable if its actions hurt someone, or if something goes wrong? To think of a more frightening scenario, might these machines rebel against humans and wish to eliminate us altogether? If yes, they represent the culmination of evolution.

As a professor of electrical engineering and computer science who works in machine learning and quantum theory, I can say that researchers are divided on whether these sorts of hyperaware machines will ever exist. There’s also debate about whether machines could or should be called “conscious” in the way we think of humans, and even some animals, as conscious. Some of the questions have to do with technology; others have to do with what consciousness actually is.

Is awareness enough?

Most computer scientists think that consciousness is a characteristic that will emerge as technology develops. Some believe that consciousness involves accepting new information, storing and retrieving old information and cognitive processing of it all into perceptions and actions. If that’s right, then one day machines will indeed be the ultimate consciousness. They’ll be able to gather more information than a human, store more than many libraries, access vast databases in milliseconds and compute all of it into decisions more complex, and yet more logical, than any person ever could.

On the other hand, there are physicists and philosophers who say there’s something more about human behavior that cannot be computed by a machine. Creativity, for example, and the sense of freedom people possess don’t appear to come from logic or calculations.

Yet these are not the only views of what consciousness is, or whether machines could ever achieve it.

Quantum views

Another viewpoint on consciousness comes from quantum theory, which is the deepest theory of physics. According to the orthodox Copenhagen Interpretation, consciousness and the physical world are complementary aspects of the same reality. When a person observes, or experiments on, some aspect of the physical world, that person’s conscious interaction causes discernible change. Since it takes consciousness as a given and no attempt is made to derive it from physics, the Copenhagen Interpretation may be called the “big-C” view of consciousness, where it is a thing that exists by itself – although it requires brains to become real. This view was popular with the pioneers of quantum theory such as Niels Bohr, Werner Heisenberg and Erwin Schrödinger.

The interaction between consciousness and matter leads to paradoxes that remain unresolved after 80 years of debate. A well-known example of this is the paradox of Schrödinger’s cat, in which a cat is placed in a situation that results in it being equally likely to survive or die – and the act of observation itself is what makes the outcome certain.

The opposing view is that consciousness emerges from biology, just as biology itself emerges from chemistry which, in turn, emerges from physics. We call this less expansive concept of consciousness “little-C.” It agrees with the neuroscientists’ view that the processes of the mind are identical to states and processes of the brain. It also agrees with a more recent interpretation of quantum theory motivated by an attempt to rid it of paradoxes, the Many Worlds Interpretation, in which observers are a part of the mathematics of physics.

Philosophers of science believe that these modern quantum physics views of consciousness have parallels in ancient philosophy. Big-C is like the theory of mind in Vedanta – in which consciousness is the fundamental basis of reality, on par with the physical universe.

Little-C, in contrast, is quite similar to Buddhism. Although the Buddha chose not to address the question of the nature of consciousness, his followers declared that mind and consciousness arise out of emptiness or nothingness.

Big-C and scientific discovery

Scientists are also exploring whether consciousness is always a computational process. Some scholars have argued that the creative moment is not at the end of a deliberate computation. For instance, dreams or visions are supposed to have inspired Elias Howe‘s 1845 design of the modern sewing machine, and August Kekulé’s discovery of the structure of benzene in 1862.

A dramatic piece of evidence in favor of big-C consciousness existing all on its own is the life of self-taught Indian mathematician Srinivasa Ramanujan, who died in 1920 at the age of 32. His notebook, which was lost and forgotten for about 50 years and published only in 1988, contains several thousand formulas, without proof in different areas of mathematics, that were well ahead of their time. Furthermore, the methods by which he found the formulas remain elusive. He himself claimed that they were revealed to him by a goddess while he was asleep.

The concept of big-C consciousness raises the questions of how it is related to matter, and how matter and mind mutually influence each other. Consciousness alone cannot make physical changes to the world, but perhaps it can change the probabilities in the evolution of quantum processes. The act of observation can freeze and even influence atoms’ movements, as Cornell physicists proved in 2015. This may very well be an explanation of how matter and mind interact.

Mind and self-organizing systems

It is possible that the phenomenon of consciousness requires a self-organizing system, like the brain’s physical structure. If so, then current machines will come up short.

Scholars don’t know if adaptive self-organizing machines can be designed to be as sophisticated as the human brain; we lack a mathematical theory of computation for systems like that. Perhaps it’s true that only biological machines can be sufficiently creative and flexible. But then that suggests people should – or soon will – start working on engineering new biological structures that are, or could become, conscious.

Reprinted with permission from The Conversation

Posted on

A breakthrough low-light image sensor for photography, life sciences, security

A sample photo (right) taken with the one-megapixel low-light Quanta Image Sensor operating at 1,040 frames per second. It is a binary single-photon image, so if the pixel was hit by one or more photons, it is white; if not, it is black. The photo was created by summing up eight frames of binary images taken continuously. A de-noising algorithm was applied to the final image. (credit: Jiaju Ma, adapted by KurzweilAI)

Engineers from Dartmouth’s Thayer School of Engineering have created a radical new imaging technology called “Quanta Image Sensor” (QIS) that may revolutionize a wide variety of imaging applications that require high quality at low light.

These include security, photography, cinematography, and medical and life sciences research.

Low-light photography (at night with only moonlight, for example) currently requires photographers to use time exposure (keeping the shutter open for seconds or minutes), making it impossible to photograph moving images.

Capturing single photons at room temperature

The new QIS technology can capture or count at the lowest possible level of light (single photons) with a resolution as high as one megapixel* (one million pixels) — scalable for higher resolution up to hundreds of megapixels per chip** — and as fast as thousands of frames*** per second (required for “bullet time” cinematography in “The Matrix”).

The QIS works at room temperature, using existing mainstream CMOS image sensor technology. Current lab-research technology may require cooling to very low temperatures, such as 4 kelvin, and is limited to low pixel count.

Quanta Image Sensor applications (credit: Gigajot)

For astrophysicists, the QIS will allow for detecting and capturing signals from distant objects in space at higher quality. For life-science researchers, it will provide improved visualization of cells under a microscope, which is critical for determining the effectiveness of therapies.

The QIS technology is commercially accessible, inexpensive, and compatible with mass-production manufacturing, according to inventor Eric R. Fossum, professor of engineering at Dartmouth. Fossum is senior author of an open-access paper on QIS in the Dec. 20 issue of The Optical Society’s (OSA) Optica. He invented the CMOS image sensor found in nearly all smartphones and cameras in the world today.

The research was performed in cooperation with Rambus, Inc. and the Taiwan Semiconductor Manufacturing Corporation and was funded by Rambus and the Defense Advanced Research Projects Agency (DARPA). The low-light capability promises to allow for improved security uses. Fossum and associates have co-founded the startup company Gigajot Technology to further develop and apply the technology to promising applications.

* By comparison, the iPhone 8 can capture 12 megapixels (but is not usable in low light).

** The technology is based on what the researchers call “jots,” which function like miniature pixels. Each jot can collect one photon, enabling the extreme low-light capability and high resolution.

*** By comparison, the iPhone 8 can record 24 to 60 frames per second.

Abstract of Photon-number-resolving megapixel image sensor at room temperature without avalanche gain

In several emerging fields of study such as encryption in optical communications, determination of the number of photons in an optical pulse is of great importance. Typically, such photon-number-resolving sensors require operation at very low temperature (e.g., 4 K for superconducting-based detectors) and are limited to low pixel count (e.g., hundreds). In this paper, a CMOS-based photon-counting image sensor is presented with photon-number-resolving capability that operates at room temperature with resolution of 1 megapixel. Termed a quanta image sensor, the device is implemented in a commercial stacked (3D) backside-illuminated CMOS image sensor process. Without the use of avalanche multiplication, the 1.1 μm pixel-pitch device achieves 0.21e−  rms0.21e−  rms average read noise with average dark count rate per pixel less than 0.2e−/s0.2e−/s, and 1040 fps readout rate. This novel platform technology fits the needs of high-speed, high-resolution, and accurate photon-counting imaging for scientific, space, security, and low-light imaging as well as a broader range of other applications.

Posted on

How to program DNA like we do computers

A programmable chemical oscillator made from DNA (credit: Ella Maru Studio and Cody Geary)

Researchers at The University of Texas at Austin have programmed DNA molecules to follow specific instructions to create sophisticated molecular machines that could be capable of communication, signal processing, problem-solving, decision-making, and control of motion in living cells — the kind of computation previously only possible with electronic circuits.

Future applications may include health care, advanced materials, and nanotechnology.

As a demonstration, the researchers constructed a first-of-its-kind chemical oscillator that uses only DNA components — no proteins, enzymes or other cellular components — to create a classic chemical reaction network (CRN) called a “rock-paper-scissors oscillator.” The goal was to show that DNA alone is capable of precise, complex behavior.

A systematic pipeline for programming DNA-only dynamical systems and the implementation of a chemical oscillator (credit: Niranjan Srinivas et al./Science)

Chemical oscillators have long been studied by engineers and scientists. For example, the researchers who discovered the chemical oscillator that controls the human circadian rhythm — responsible for our bodies’ day and night rhythm — earned the 2017 Nobel Prize in physiology or medicine.

“As engineers, we are very good at building sophisticated electronics, but biology uses complex chemical reactions inside cells to do many of the same kinds of things, like making decisions,” said David Soloveichik, an assistant professor in the Cockrell School’s Department of Electrical and Computer Engineering and senior author of a paper in the journal Science.

“Eventually, we want to be able to interact with the chemical circuits of a cell, or fix malfunctioning circuits or even reprogram them for greater control. But in the near term, our DNA circuits could be used to program the behavior of cell-free chemical systems that synthesize complex molecules, diagnose complex chemical signatures, and respond to their environments.”

The team’s research was conducted as part of the National Science Foundation’s (NSF) Molecular Programming Project and funded by the NSF, the Office of Naval Research, the National Institutes of Health, and the Gordon and Betty Moore Foundation.


Programming a Chemical Oscillator

Abstract of Enzyme-free nucleic acid dynamical systems

An important goal of synthetic biology is to create biochemical control systems with the desired characteristics from scratch. Srinivas et al. describe the creation of a biochemical oscillator that requires no enzymes or evolved components, but rather is implemented through DNA molecules designed to function in strand displacement cascades. Furthermore, they created a compiler that could translate a formal chemical reaction network into the necessary DNA sequences that could function together to provide a specified dynamic behavior.

 

 

Posted on

A new low-cost, simple way to measure medical vital signs with radio waves

A radio-frequency identification (RFID) tag, used to monitor vital signs, can go into your pocket or be woven into a shirt (credit: Cornell)

Replacing devices based on 19th-century technology* and still in use, Cornell University engineers have developed a simple method for gathering blood pressure, heart rate, and breath rate from multiple patients simultaneously. It uses low-power radio-frequency signals and low-cost microchip radio-frequency identification (RFID) “tags” — similar to the ubiquitous anti-theft tags used in department stores.

The RFID tags measure internal body motion, such as a heart as it beats or blood as it pulses under skin. Powered remotely by electromagnetic energy supplied by a central reader, the tags use a new concept called “near-field coherent sensing.” Mechanical motions (heartbeat, etc.) in the body modulate (modify) radio waves that are bounced off the body and internal organs by passive (no battery required) RFID tags.

The modulated signals detected by the tag then bounce back to an electronic reader, located elsewhere in the room, that gathers the data. Each tag has a unique identification code that it transmits with its signal, allowing up to 200 people to be monitored simultaneously.

Electromagnetic simulations of monitoring vital signs via radio transmission, showing heartbeat sensing (left) and pulse sensing (right) (credit: Xiaonan Hui and Edwin C. Kan/Nature Electronics)

“If this is an emergency room, everybody that comes in can wear these tags or can simply put tags in their front pockets, and everybody’s vital signs can be monitored at the same time. I’ll know exactly which person each of the vital signs belongs to,” said Edwin Kan, a Cornell professor of electrical and computer engineering.

The signal is as accurate as an electrocardiogram or a blood-pressure cuff, according to Kan, who believes the technology could also be used to measure bowel movement, eye movement, and many other internal mechanical motions produced by the body.

The researchers envision embedding the RFID chips in clothing to monitor health in real time, with little or no effort required by the user. They have also developed a method for embroidering the tags directly onto clothing using fibers coated with nanoparticles. A cellphone could read (and display) your vital signs and also transmit them for remote medical monitoring.

The system is detailed in the open-access paper “Monitoring Vital Signs Over Multiplexed Radio by Near-Field Coherent Sensing,” published online Nov. 27 in the journal Nature Electronics. “Current approaches to monitoring vital signs are based on body electrodes, optical absorption, pressure or strain gauges, stethoscope, and ultrasound or radiofrequency (RF) backscattering, each of which suffers particular drawbacks during application,” the paper notes.

* The sphygmomanometer was invented by Samuel Siegfried Karl Ritter von Basch in 1881. Devices based on its basic pressure principle are still in use. (credit: Wellcome Trustees)

Abstract of Monitoring vital signs over multiplexed radio by near-field coherent sensing

Monitoring the heart rate, blood pressure, respiration rate and breath effort of a patient is critical to managing their care, but current approaches are limited in terms of sensing capabilities and sampling rates. The measurement process can also be uncomfortable due to the need for direct skin contact, which can disrupt the circadian rhythm and restrict the motion of the patient. Here we show that the external and internal mechanical motion of a person can be directly modulated onto multiplexed radiofrequency signals integrated with unique digital identification using near-field coherent sensing. The approach, which does not require direct skin contact, offers two possible implementations: passive and active radiofrequency identification tags. To minimize deployment and maintenance cost, passive tags can be integrated into garments at the chest and wrist areas, where the two multiplexed far-field backscattering waveforms are collected at the reader to retrieve the heart rate, blood pressure, respiration rate and breath effort. To maximize reading range and immunity to multipath interference caused by indoor occupant motion, active tags could be placed in the front pocket and in the wrist cuff to measure the antenna reflection due to near-field coherent sensing and then the vital signals sampled and transmitted entirely in digital format. Our system is capable of monitoring multiple people simultaneously and could lead to the cost-effective automation of vital sign monitoring in care facilities.

Posted on

Video games and piano lessons improve cognitive functions in seniors, researchers find

(credit: Nintendo)

For seniors, playing 3D-platform games like Super Mario 64 or taking piano lessons can stave off mild cognitive impairment and perhaps even prevent Alzheimer’s disease, according to a new study by Université de Montréal psychology professors.

In the studies, 33 people ages 55 to 75 were instructed to play Super Mario 64 for 30 minutes a day, five days a week for a period of six months, or take piano lessons (for the first time in their life) with the same frequency and in the same sequence. A control group did not perform any particular task.

The researchers evaluated the effects of the experiment with cognitive performance tests and magnetic resonance imaging (MRI) to measure variations in the volume of gray matter.

Increased gray matter in the left and right hippocampus and the left cerebellum after older adults completed six months of video-game training. (credit: Greg L. West et al./PLoS One)

  • The participants in the video-game cohort saw increases in gray matter volume in the cerebellum (plays a major role in motor control and balance) and the hippocampus (associated with spatial and episodic memory, a key factor in long-term cognitive health); and their short-term memory improved. (The hippocampus gray matter acts as a marker for neurological disorders that can occur over time, including mild cognitive impairment and Alzheimer’s.)
  • There were gray-matter increases in the dorsolateral prefrontal cortex (controls planning, decision-making, and inhibition) and cerebellum of the participants who took piano lessons.
  • Some degree of atrophy was noted in all three areas of the brain among those in the passive control group.

“These findings can also be used to drive future research on Alzheimer’s, since there is a link between the volume of the hippocampus and the risk of developing the disease,” said Gregory West, an associate professor at the Université de Montréal and lead author of an open-access paper in PLoS One journal.

“3-D video games engage the hippocampus into creating a cognitive map, or a mental representation, of the virtual environment that the brain is exploring,” said West. “Several studies suggest stimulation of the hippocampus increases both functional activity and gray matter within this region.”

However, “It remains to be seen whether it is specifically brain activity associated with spatial memory that affects plasticity, or whether it’s simply a matter of learning something new.”

Researchers at the Memorial University in Newfoundland, and Montreal’s Douglas Hospital Research Centre were also involved in the study.

In two previous studies by the researchers in 2014 and 2017, young adults in their twenties were asked to play 3D video games of logic and puzzles on platforms like Super Mario 64. Findings showed that the gray matter in their hippocampus also increased after training.

Abstract of Playing Super Mario 64 increases hippocampal grey matter in older adults

Maintaining grey matter within the hippocampus is important for healthy cognition. Playing 3D-platform video games has previously been shown to promote grey matter in the hippocampus in younger adults. In the current study, we tested the impact of 3D-platform video game training (i.e., Super Mario 64) on grey matter in the hippocampus, cerebellum, and the dorsolateral prefrontal cortex (DLPFC) of older adults. Older adults who were 55 to 75 years of age were randomized into three groups. The video game experimental group (VID; n = 8) engaged in a 3D-platform video game training over a period of 6 months. Additionally, an active control group took a series of self-directed, computerized music (piano) lessons (MUS; n = 12), while a no-contact control group did not engage in any intervention (CON; n = 13). After training, a within-subject increase in grey matter within the hippocampus was significant only in the VID training group, replicating results observed in younger adults. Active control MUS training did, however, lead to a within-subject increase in the DLPFC, while both the VID and MUS training produced growth in the cerebellum. In contrast, the CON group displayed significant grey matter loss in the hippocampus, cerebellum and the DLPFC.

Posted on

AlphaZero’s ‘alien’ superhuman-level program masters chess in 24 hours with no domain knowledge

AlphaZero vs. Stockfish chess program | Round 1 (credit: Chess.com)

Demis Hassabis, the founder and CEO of DeepMind, announced at the Neural Information Processing Systems conference (NIPS 2017) last week that DeepMind’s new AlphaZero program achieved a superhuman level of play in chess within 24 hours.

The program started from random play, given no domain knowledge except the game rules, according to an arXiv paper by DeepMind researchers published Dec. 5.

“It doesn’t play like a human, and it doesn’t play like a program,” said Hassabis, an expert chess player himself. “It plays in a third, almost alien, way. It’s like chess from another dimension.”

AlphaZero also mastered both shogi (Japanese chess) and Go within 24 hours, defeating a world-champion program in all three cases. The original AlphaGo mastered Go by learning thousands of example games and then practicing against another version of itself.

“AlphaZero was not ‘taught’ the game in the traditional sense,” explains Chess.com. “That means no opening book, no endgame tables, and apparently no complicated algorithms dissecting minute differences between center pawns and side pawns. This would be akin to a robot being given access to thousands of metal bits and parts, but no knowledge of a combustion engine, then it experiments numerous times with every combination possible until it builds a Ferrari. … The program had four hours to play itself many, many times, thereby becoming its own teacher.”

“What’s also remarkable, though, Hassabis explained, is that it sometimes makes seemingly crazy sacrifices, like offering up a bishop and queen to exploit a positional advantage that led to victory,” MIT Technology Review notes. “Such sacrifices of high-value pieces are normally rare. In another case the program moved its queen to the corner of the board, a very bizarre trick with a surprising positional value.”

Abstract of Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

The game of chess is the most widely-studied domain in the history of artificial intelligence. The strongest programs are based on a combination of sophisticated search techniques, domain-specific adaptations, and handcrafted evaluation functions that have been refined by human experts over several decades. In contrast, the AlphaGo Zero program recently achieved superhuman performance in the game of Go, by tabula rasa reinforcement learning from games of self-play. In this paper, we generalise this approach into a single AlphaZero algorithm that can achieve, tabula rasa, superhuman performance in many challenging domains. Starting from random play, and given no domain knowledge except the game rules, AlphaZero achieved within 24 hours a superhuman level of play in the games of chess and shogi (Japanese chess) as well as Go, and convincingly defeated a world-champion program in each case.

Posted on

3D-printing biocompatible living bacteria

3D-printing with an ink containing living bacteria (credit: Bara Krautz/bara@scienceanimated.com)

Researchers at ETH Zurich university have developed a technique for 3D-printing biocompatible living bacteria for the first time — making it possible to produce produce high-purity cellulose for biomedical applications and nanofilters that can break down toxic substances (in drinking water, for example) or for use in disastrous oil spills, for example.

The technique, called “Flink” (“functional living ink”) allows for printing mini biochemical factories with properties that vary based on which species of bacteria are used. Up to four different inks containing different species of bacteria at different concentrations can be printed in a single pass.

Schematics of the Flink 3D bacteria-printing process for creating two types of functional living materials. (Left and center) Bacteria are embedded in a biocompatible hydrogel (which provides the supporting structure). (Right) The inclusion of P. putida* or A. xylinum* bacteria in the ink yields 3D-printed materials capable of degrading environmental pollutants (top) or forming bacterial cellulose in situ for biomedical applications (bottom), respectively. (credit: Manuel Schaffner et al./Science Advances)

The technique was described Dec. 1, 2017 in the open-access journal Science Advances.

(Left) A. xylinum bacteria were used in printing a cellulose nanofibril network (scanning electron microscope image), which was deposited (Right) on a doll face, forming a cellulose-reinforced hydrogel that, after removal of all biological residues, could serve as a skin transplant. (credit: Manuel Schaffner et al./Science Advances)

“The in situ formation of reinforcing cellulose fibers within the hydrogel is particularly attractive for regions under mechanical tension, such as the elbow and knee, or when administered as a pouch onto organs to prevent fibrosis after surgical implants and transplantations,” the researchers note in the paper. “Cellulose films grown in complex geometries precisely match the topography of the site of interest, preventing the formation of wrinkles and entrapments of contaminants that could impair the healing process. We envision that long-term medical applications will benefit from the presented multimaterial 3D printing process by locally deploying bacteria where needed.”

 * Pseudomonas putida breaks down the toxic chemical phenol, which is produced on a grand scale in the chemical industry; Acetobacter xylinum secretes high-purity nanocellulose, which relieves pain, retains moisture and is stable, opening up potential applications in the treatment of burns.

Abstract of 3D printing of bacteria into functional complex materials

Despite recent advances to control the spatial composition and dynamic functionalities of bacteria embedded in materials, bacterial localization into complex three-dimensional (3D) geometries remains a major challenge. We demonstrate a 3D printing approach to create bacteria-derived functional materials by combining the natural diverse metabolism of bacteria with the shape design freedom of additive manufacturing. To achieve this, we embedded bacteria in a biocompatible and functionalized 3D printing ink and printed two types of “living materials” capable of degrading pollutants and of producing medically relevant bacterial cellulose. With this versatile bacteria-printing platform, complex materials displaying spatially specific compositions, geometry, and properties not accessed by standard technologies can be assembled from bottom up for new biotechnological and biomedical applications.

Posted on

New technology allows robots to visualize their own future


UC Berkeley | Vestri the robot imagines how to perform tasks

UC Berkeley researchers have developed a robotic learning technology that enables robots to imagine the future of their actions so they can figure out how to manipulate objects they have never encountered before. It could help self-driving cars anticipate future events on the road and produce more intelligent robotic assistants in homes.

The initial prototype focuses on learning simple manual skills entirely from autonomous play — similar to how children can learn about their world by playing with toys, moving them around, grasping, etc.

Using this technology, called visual foresight, the robots can predict what their cameras will see if they perform a particular sequence of movements. These robotic imaginations are still relatively simple for now — predictions made only several seconds into the future — but they are enough for the robot to figure out how to move objects around on a table without disturbing obstacles.

The robot can learn to perform these tasks without any help from humans or prior knowledge about physics, its environment, or what the objects are. That’s because the visual imagination is learned entirely from scratch from unattended and unsupervised (no humans involved) exploration, where the robot plays with objects on a table.

After this play phase, the robot builds a predictive model of the world, and can use this model to manipulate new objects that it has not seen before.

“In the same way that we can imagine how our actions will move the objects in our environment, this method can enable a robot to visualize how different behaviors will affect the world around it,” said Sergey Levine, assistant professor in Berkeley’s Department of Electrical Engineering and Computer Sciences, whose lab developed the technology. “This can enable intelligent planning of highly flexible skills in complex real-world situations.”

The research team demonstrated the visual foresight technology at the Neural Information Processing Systems conference in Long Beach, California, on Monday, December 4, 2017.

Learning by playing: how it works

Robot’s imagined predictions (credit: UC Berkeley)

At the core of this system is a deep learning technology based on convolutional recurrent video prediction, or dynamic neural advection (DNA). DNA-based models predict how pixels in an image will move from one frame to the next, based on the robot’s actions. Recent improvements to this class of models, as well as greatly improved planning capabilities, have enabled robotic control based on video prediction to perform increasingly complex tasks, such as sliding toys around obstacles and repositioning multiple objects.

“In that past, robots have learned skills with a human supervisor helping and providing feedback. What makes this work exciting is that the robots can learn a range of visual object manipulation skills entirely on their own,” said Chelsea Finn, a doctoral student in Levine’s lab and inventor of the original DNA model.

With the new technology, a robot pushes objects on a table, then uses the learned prediction model to choose motions that will move an object to a desired location. Robots use the learned model from raw camera observations to teach themselves how to avoid obstacles and push objects around obstructions.

Since control through video prediction relies only on observations that can be collected autonomously by the robot, such as through camera images, the resulting method is general and broadly applicable. Building video prediction models only requires unannotated video, which can be collected by the robot entirely autonomously.

That contrasts with conventional computer-vision methods, which require humans to manually label thousands or even millions of images.

Posted on

Why (most) future robots won’t look like robots

A future robot’s body could combine soft actuators and stiff structure, with distributed computation throughout — an example of the new “material robotics.” (credit: Nikolaus Correll/University of Colorado)

Future robots won’t be limited to humanoid form (like Boston Robotics’ formidable backflipping Atlas). They’ll be invisibly embedded everywhere in common objects.

Such as a shoe that can intelligently support your gait, change stiffness as you’re running or walking, and adapt to different surfaces — or even help you do backflips.

That’s the vision of researchers at Oregon State University, the University of Colorado, Yale University, and École Polytechnique Fédérale de Lausanne, who describe the burgeoning new field of  “material robotics” in a perspective article published Nov. 29, 2017 in Science Robotics. (The article cites nine articles in this special issue, three of which you can access for free.)

Disappearing into the background of everyday life

The authors challenge a widespread basic assumption: that robots are either “machines that run bits of code” or “software ‘bots’ interacting with the world through a physical instrument.”

“We take a third path: one that imbues intelligence into the very matter of a robot,” says Oregon State University researcher Yiğit Mengüç, an assistant professor of mechanical engineering in OSU’s College of Engineering and part of the college’s Collaborative Robotics and Intelligent Systems Institute.

On that path, materials scientists are developing new bulk materials with the inherent multifunctionality required for robotic applications, while roboticists are working on new material systems with tightly integrated components, disappearing into the background of everyday life. “The spectrum of possible ap­proaches spans from soft grippers with zero knowledge and zero feedback all the way to humanoids with full knowledge and full feed­back,” the authors note in the paper.

For example, “In the future, your smartphone may be made from stretchable, foldable material so there’s no danger of it shattering,” says Mengüç. “Or it might have some actuation, where it changes shape in your hand to help with the display, or it can be able to communicate something about what you’re observing on the screen. All these bits and pieces of technology that we take for granted in life will be living, physically responsive things, moving, changing shape in response to our needs, not just flat, static screens.”

Soft robots get superpowers

Origami-inspired artificial muscles capable of lifting up to 1,000 times their own weight, simply by applying air or water pressure (credit: Shuguang Li/Wyss Institute at Harvard University)

As a good example of material-enabled robotics, researchers at the Wyss Institute at Harvard University and MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed origami-inspired, programmable, super-strong artificial muscles that will allow future soft robots to lift objects that are up to 1,000 times their own weight — using only air or water pressure.

The actuators are “programmed” by the structural design itself. The skeleton folds define how the whole structure moves — no control system required.

That allows the muscles to be very compact and simple, which makes them more appropriate for mobile or body-mounted systems that can’t accommodate large or heavy machinery, says Shuguang Li, Ph.D., a Postdoctoral Fellow at the Wyss Institute and MIT CSAIL and first author of an an open-access article on the research published Nov. 21, 2017 in Proceedings of the National Academy of Sciences (PNAS).

Each artificial muscle consists of an inner “skeleton” that can be made of various materials, such as a metal coil or a sheet of plastic folded into a certain pattern, surrounded by air or fluid and sealed inside a plastic or textile bag that serves as the “skin.” The structural geometry of the skeleton itself determines the muscle’s motion. A vacuum applied to the inside of the bag initiates the muscle’s movement by causing the skin to collapse onto the skeleton, creating tension that drives the motion. Incredibly, no other power source or human input is required to direct the muscle’s movement — it’s automagically determined entirely by the shape and composition of the skeleton. (credit: Shuguang Li/Wyss Institute at Harvard University)

Resilient, multipurpose, scalable

Not only can the artificial muscles move in many ways, they do so with impressive resilience. They can generate about six times more force per unit area than mammalian skeletal muscle can, and are also incredibly lightweight. A 2.6-gram muscle can lift a 3-kilogram object, which is the equivalent of a mallard duck lifting a car. Additionally, a single muscle can be constructed within ten minutes using materials that cost less than $1, making them cheap and easy to test and iterate.

These muscles can be powered by a vacuum, which makes them safer than most of the other artificial muscles currently being tested. The muscles have been built in sizes ranging from a few millimeters up to a meter. So the muscles can be used in numerous applications at multiple scales, from miniature surgical devices to wearable robotic exoskeletons, transformable architecture, and deep-sea manipulators for research or construction, up to large deployable structures for space exploration.

The team could also construct the muscles out of the water-soluble polymer PVA. That opens the possibility of bio-friendly robots that can perform tasks in natural settings with minimal environmental impact, or ingestible robots that move to the proper place in the body and then dissolve to release a drug.

The team constructed dozens of muscles using materials ranging from metal springs to packing foam to sheets of plastic, and experimented with different skeleton shapes to create muscles that can contract down to 10% of their original size, lift a delicate flower off the ground, and twist into a coil, all simply by sucking the air out of them.

This research was funded by the Defense Advanced Research Projects Agency (DARPA), the National Science Foundation (NSF), and the Wyss Institute for Biologically Inspired Engineering.


Wyss Institute | Origami-Inspired Artificial Muscles

Abstract of Fluid-driven origami-inspired artificial muscles

Artificial muscles hold promise for safe and powerful actuation for myriad common machines and robots. However, the design, fabrication, and implementation of artificial muscles are often limited by their material costs, operating principle, scalability, and single-degree-of-freedom contractile actuation motions. Here we propose an architecture for fluid-driven origami-inspired artificial muscles. This concept requires only a compressible skeleton, a flexible skin, and a fluid medium. A mechanical model is developed to explain the interaction of the three components. A fabrication method is introduced to rapidly manufacture low-cost artificial muscles using various materials and at multiple scales. The artificial muscles can be programed to achieve multiaxial motions including contraction, bending, and torsion. These motions can be aggregated into systems with multiple degrees of freedom, which are able to produce controllable motions at different rates. Our artificial muscles can be driven by fluids at negative pressures (relative to ambient). This feature makes actuation safer than most other fluidic artificial muscles that operate with positive pressures. Experiments reveal that these muscles can contract over 90% of their initial lengths, generate stresses of ∼600 kPa, and produce peak power densities over 2 kW/kg—all equal to, or in excess of, natural muscle. This architecture for artificial muscles opens the door to rapid design and low-cost fabrication of actuation systems for numerous applications at multiple scales, ranging from miniature medical devices to wearable robotic exoskeletons to large deployable structures for space exploration.

Posted on

Using light instead of electrons promises faster, smaller, more-efficient computers and smartphones

Trapped light for optical computation (credit: Imperial College London)

By forcing light to go through a smaller gap than ever before, a research team at Imperial College London has taken a step toward computers based on light instead of electrons.

Light would be preferable for computing because it can carry much-higher-density information, it’s much faster, and more efficient (generates little to no heat). But light beams don’t easily interact with one other. So information on high-speed fiber-optic cables (provided by your cable TV company, for example) currently has to be converted (via a modem or other device) into slower signals (electrons on wires or wireless signals) to allow for processing the data on devices such as computers and smartphones.

Electron-microscope image of an optical-computing nanofocusing device that is 25 nanometers wide and 2 micrometers long, using grating couplers (vertical lines) to interface with fiber-optic cables. (credit: Nielsen et al., 2017/Imperial College London)

To overcome that limitation, the researchers used metamaterials to squeeze light into a metal channel only 25 nanometers (billionths of a meter) wide, increasing its intensity and allowing photons to interact over the range of micrometers (millionths of meters) instead of centimeters.*

That means optical computation that previously required a centimeters-size device can now be realized on the micrometer (one millionth of a meter) scale, bringing optical processing into the size range of electronic transistors.

The results were published Thursday Nov. 30, 2017 in the journal Science.

* Normally, when two light beams cross each other, the individual photons do not interact or alter each other, as two electrons do when they meet. That means a long span of material is needed to gradually accumulate the effect and make it useful. Here, a “plasmonic nanofocusing” waveguide is used, strongly confining light within a nonlinear organic polymer.

Abstract of Giant nonlinear response at a plasmonic nanofocus drives efficient four-wave mixing

Efficient optical frequency mixing typically must accumulate over large interaction lengths because nonlinear responses in natural materials are inherently weak. This limits the efficiency of mixing processes owing to the requirement of phase matching. Here, we report efficient four-wave mixing (FWM) over micrometer-scale interaction lengths at telecommunications wavelengths on silicon. We used an integrated plasmonic gap waveguide that strongly confines light within a nonlinear organic polymer. The gap waveguide intensifies light by nanofocusing it to a mode cross-section of a few tens of nanometers, thus generating a nonlinear response so strong that efficient FWM accumulates over wavelength-scale distances. This technique opens up nonlinear optics to a regime of relaxed phase matching, with the possibility of compact, broadband, and efficient frequency mixing integrated with silicon photonics.