Tag Archives: current
#437769 Q&A: Facebook’s CTO Is at War With ...
Photo: Patricia de Melo Moreira/AFP/Getty Images
Facebook chief technology officer Mike Schroepfer leads the company’s AI and integrity efforts.
Facebook’s challenge is huge. Billions of pieces of content—short and long posts, images, and combinations of the two—are uploaded to the site daily from around the world. And any tiny piece of that—any phrase, image, or video—could contain so-called bad content.
In its early days, Facebook relied on simple computer filters to identify potentially problematic posts by their words, such as those containing profanity. These automatically filtered posts, as well as posts flagged by users as offensive, went to humans for adjudication.
In 2015, Facebook started using artificial intelligence to cull images that contained nudity, illegal goods, and other prohibited content; those images identified as possibly problematic were sent to humans for further review.
By 2016, more offensive photos were reported by Facebook’s AI systems than by Facebook users (and that is still the case).
In 2018, Facebook CEO Mark Zuckerberg made a bold proclamation: He predicted that within five or ten years, Facebook’s AI would not only look for profanity, nudity, and other obvious violations of Facebook’s policies. The tools would also be able to spot bullying, hate speech, and other misuse of the platform, and put an immediate end to them.
Today, automated systems using algorithms developed with AI scan every piece of content between the time when a user completes a post and when it is visible to others on the site—just fractions of a second. In most cases, a violation of Facebook’s standards is clear, and the AI system automatically blocks the post. In other cases, the post goes to human reviewers for a final decision, a workforce that includes 15,000 content reviewers and another 20,000 employees focused on safety and security, operating out of more than 20 facilities around the world.
In the first quarter of this year, Facebook removed or took other action (like appending a warning label) on more than 9.6 million posts involving hate speech, 8.6 million involving child nudity or exploitation, almost 8 million posts involving the sale of drugs, 2.3 million posts involving bullying and harassment, and tens of millions of posts violating other Facebook rules.
Right now, Facebook has more than 1,000 engineers working on further developing and implementing what the company calls “integrity” tools. Using these systems to screen every post that goes up on Facebook, and doing so in milliseconds, is sucking up computing resources. Facebook chief technology officer Mike Schroepfer, who is heading up Facebook’s AI and integrity efforts, spoke with IEEE Spectrum about the team’s progress on building an AI system that detects bad content.
Since that discussion, Facebook’s policies around hate speech have come under increasing scrutiny, with particular attention on divisive posts by political figures. A group of major advertisers in June announced that they would stop advertising on the platform while reviewing the situation, and civil rights groups are putting pressure on others to follow suit until Facebook makes policy changes related to hate speech and groups that promote hate, misinformation, and conspiracies.
Facebook CEO Mark Zuckerberg responded with news that Facebook will widen the category of what it considers hateful content in ads. Now the company prohibits claims that people from a specific race, ethnicity, national origin, religious affiliation, caste, sexual orientation, gender identity, or immigration status are a threat to the physical safety, health, or survival of others. The policy change also aims to better protect immigrants, migrants, refugees, and asylum seekers from ads suggesting these groups are inferior or expressing contempt. Finally, Zuckerberg announced that the company will label some problematic posts by politicians and government officials as content that violates Facebook’s policies.
However, civil rights groups say that’s not enough. And an independent audit released in July also said that Facebook needs to go much further in addressing civil rights concerns and disinformation.
Schroepfer indicated that Facebook’s AI systems are designed to quickly adapt to changes in policy. “I don’t expect considerable technical changes are needed to adjust,” he told Spectrum.
This interview has been edited and condensed for clarity.
IEEE Spectrum: What are the stakes of content moderation? Is this an existential threat to Facebook? And is it critical that you deal well with the issue of election interference this year?
Schroepfer: It’s probably existential; it’s certainly massive. We are devoting a tremendous amount of our attention to it.
The idea that anyone could meddle in an election is deeply disturbing and offensive to all of us here, just as people and citizens of democracies. We don’t want to see that happen anywhere, and certainly not on our watch. So whether it’s important to the company or not, it’s important to us as people. And I feel a similar way on the content-moderation side.
There are not a lot of easy choices here. The only way to prevent people, with certainty, from posting bad things is to not let them post anything. We can take away all voice and just say, “Sorry, the Internet’s too dangerous. No one can use it.” That will certainly get rid of all hate speech online. But I don’t want to end up in that world. And there are variants of that world that various governments are trying to implement, where they get to decide what’s true or not, and you as a person don’t. I don’t want to get there either.
My hope is that we can build a set of tools that make it practical for us to do a good enough job, so that everyone is still excited about the idea that anyone can share what they want, and so that Facebook is a safe and reasonable place for people to operate in.
Spectrum: You joined Facebook in 2008, before AI was part of the company’s toolbox. When did that change? When did you begin to think that AI tools would be useful to Facebook?
Schroepfer: Ten years ago, AI wasn’t commercially practical; the technology just didn’t work very well. In 2012, there was one of those moments that a lot of people point to as the beginning of the current revolution in deep learning and AI. A computer-vision model—a neural network—was trained using what we call supervised training, and it turned out to be better than all the existing models.
Spectrum: How is that training done, and how did computer-vision models come to Facebook?
Image: Facebook
Just Broccoli? Facebook’s image analysis algorithms can tell the difference between marijuana [left] and tempura broccoli [right] better than some humans.
Schroepfer: Say I take a bunch of photos and I have people look at them. If they see a photo of a cat, they put a text label that says cat; if it’s one of a dog, the text label says dog. If you build a big enough data set and feed that to the neural net, it learns how to tell the difference between cats and dogs.
Prior to 2012, it didn’t work very well. And then in 2012, there was this moment where it seemed like, “Oh wow, this technique might work.” And a few years later we were deploying that form of technology to help us detect problematic imagery.
Spectrum: Do your AI systems work equally well on all types of prohibited content?
Schroepfer: Nudity was technically easiest. I don’t need to understand language or culture to understand that this is either a naked human or not. Violence is a much more nuanced problem, so it was harder technically to get it right. And with hate speech, not only do you have to understand the language, it may be very contextual, even tied to recent events. A week before the Christchurch shooting [New Zealand, 2019], saying “I wish you were in the mosque” probably doesn’t mean anything. A week after, that might be a terrible thing to say.
Spectrum: How much progress have you made on hate speech?
Schroepfer: AI, in the first quarter of 2020, proactively detected 88.8 percent of the hate-speech content we removed, up from 80.2 percent in the previous quarter. In the first quarter of 2020, we took action on 9.6 million pieces of content for violating our hate-speech policies.
Image: Facebook
Off Label: Sometimes image analysis isn’t enough to determine whether a picture posted violates the company’s policies. In considering these candy-colored vials of marijuana, for example, the algorithms can look at any accompanying text and, if necessary, comments on the post.
Spectrum: It sounds like you’ve expanded beyond tools that analyze images and are also using AI tools that analyze text.
Schroepfer: AI started off as very siloed. People worked on language, people worked on computer vision, people worked on video. We’ve put these things together—in production, not just as research—into multimodal classifiers.
[Schroepfer shows a photo of a pan of Rice Krispies treats, with text referring to it as a “potent batch”] This is a case in which you have an image, and then you have the text on the post. This looks like Rice Krispies. On its own, this image is fine. You put the text together with it in a bigger model; that can then understand what’s going on. That didn’t work five years ago.
Spectrum: Today, every post that goes up on Facebook is immediately checked by automated systems. Can you explain that process?
Image: Facebook
Bigger Picture: Identifying hate speech is often a matter of context. Either the text or the photo in this post isn’t hateful standing alone, but putting them together tells a different story.
Schroepfer: You upload an image and you write some text underneath it, and the systems look at both the image and the text to try to see which, if any, policies it violates. Those decisions are based on our Community Standards. It will also look at other signals on the posts, like the comments people make.
It happens relatively instantly, though there may be times things happen after the fact. Maybe you uploaded a post that had misinformation in it, and at the time you uploaded it, we didn’t know it was misinformation. The next day we fact-check something and scan again; we may find your post and take it down. As we learn new things, we’re going to go back through and look for violations of what we now know to be a problem. Or, as people comment on your post, we might update our understanding of it. If people are saying, “That’s terrible,” or “That’s mean,” or “That looks fake,” those comments may be an interesting signal.
Spectrum: How is Facebook applying its AI tools to the problem of election interference?
Schroepfer: I would split election interference into two categories. There are times when you’re going after the content, and there are times you’re going after the behavior or the authenticity of the person.
On content, if you’re sharing misinformation, saying, “It’s super Wednesday, not super Tuesday, come vote on Wednesday,” that’s a problem whether you’re an American sitting in California or a foreign actor.
Other times, people create a series of Facebook pages pretending they’re Americans, but they’re really a foreign entity. That is a problem on its own, even if all the content they’re sharing completely meets our Community Standards. The problem there is that you have a foreign government running an information operation.
There, you need different tools. What you’re trying to do is put pieces together, to say, “Wait a second. All of these pages—Martians for Justice, Moonlings for Justice, and Venusians for Justice”—are all run by an administrator with an IP address that’s outside the United States. So they’re all connected, even though they’re pretending to not be connected. That’s a very different problem than me sitting in my office in Menlo Park [Calif.] sharing misinformation.
I’m not going to go into lots of technical detail, because this is an area of adversarial nature. The fundamental problem you’re trying to solve is that there’s one entity coordinating the activity of a bunch of things that look like they’re not all one thing. So this is a series of Instagram accounts, or a series of Facebook pages, or a series of WhatsApp accounts, and they’re pretending to be totally different things. We’re looking for signals that these things are related in some way. And we’re looking through the graph [what Facebook calls its map of relationships between users] to understand the properties of this network.
Spectrum: What cutting-edge AI tools and methods have you been working on lately?
Schroepfer: Supervised learning, with humans setting up the instruction process for the AI systems, is amazingly effective. But it has a very obvious flaw: the speed at which you can develop these things is limited by how fast you can curate the data sets. If you’re dealing in a problem domain where things change rapidly, you have to rebuild a new data set and retrain the whole thing.
Self-supervision is inspired by the way people learn, by the way kids explore the world around them. To get computers to do it themselves, we take a bunch of raw data and build a way for the computer to construct its own tests. For language, you scan a bunch of Web pages, and the computer builds a test where it takes a sentence, eliminates one of the words, and figures out how to predict what word belongs there. And because it created the test, it actually knows the answer. I can use as much raw text as I can find and store because it’s processing everything itself and doesn’t require us to sit down and build the information set. In the last two years there has been a revolution in language understanding as a result of AI self-supervised learning.
Spectrum: What else are you excited about?
Schroepfer: What we’ve been working on over the last few years is multilingual understanding. Usually, when I’m trying to figure out, say, whether something is hate speech or not I have to go through the whole process of training the model in every language. I have to do that one time for every language. When you make a post, the first thing we have to figure out is what language your post is in. “Ah, that’s Spanish. So send it to the Spanish hate-speech model.”
We’ve started to build a multilingual model—one box where you can feed in text in 40 different languages and it determines whether it’s hate speech or not. This is way more effective and easier to deploy.
To geek out for a second, just the idea that you can build a model that understands a concept in multiple languages at once is crazy cool. And it not only works for hate speech, it works for a variety of things.
When we started working on this multilingual model years ago, it performed worse than every single individual model. Now, it not only works as well as the English model, but when you get to the languages where you don’t have enough data, it’s so much better. This rapid progress is very exciting.
Spectrum: How do you move new AI tools from your research labs into operational use?
Schroepfer: Engineers trying to make the next breakthrough will often say, “Cool, I’ve got a new thing and it achieved state-of-the-art results on machine translation.” And we say, “Great. How long does it take to run in production?” They say, “Well, it takes 10 seconds for every sentence to run on a CPU.” And we say, “It’ll eat our whole data center if we deploy that.” So we take that state-of-the-art model and we make it 10 or a hundred or a thousand times more efficient, maybe at the cost of a little bit of accuracy. So it’s not as good as the state-of-the-art version, but it’s something we can actually put into our data centers and run in production.
Spectrum: What’s the role of the humans in the loop? Is it true that Facebook currently employs 35,000 moderators?
Schroepfer: Yes. Right now our goal is not to reduce that. Our goal is to do a better job catching bad content. People often think that the end state will be a fully automated system. I don’t see that world coming anytime soon.
As automated systems get more sophisticated, they take more and more of the grunt work away, freeing up the humans to work on the really gnarly stuff where you have to spend an hour researching.
We also use AI to give our human moderators power tools. Say I spot this new meme that is telling everyone to vote on Wednesday rather than Tuesday. I have a tool in front of me that says, “Find variants of that throughout the system. Find every photo with the same text, find every video that mentions this thing and kill it in one shot.” Rather than, I found this one picture, but then a bunch of other people upload that misinformation in different forms.
Another important aspect of AI is that anything I can do to prevent a person from having to look at terrible things is time well spent. Whether it’s a person employed by us as a moderator or a user of our services, looking at these things is a terrible experience. If I can build systems that take the worst of the worst, the really graphic violence, and deal with that in an automated fashion, that’s worth a lot to me. Continue reading →
Posted in Human Robots
#437751 Startup and Academics Find Path to ...
Engineers have been chasing a form of AI that could drastically lower the energy required to do typical AI things like recognize words and images. This analog form of machine learning does one of the key mathematical operations of neural networks using the physics of a circuit instead of digital logic. But one of the main things limiting this approach is that deep learning’s training algorithm, back propagation, has to be done by GPUs or other separate digital systems.
Now University of Montreal AI expert Yoshua Bengio, his student Benjamin Scellier, and colleagues at startup Rain Neuromorphics have come up with way for analog AIs to train themselves. That method, called equilibrium propagation, could lead to continuously learning, low-power analog systems of a far greater computational ability than most in the industry now consider possible, according to Rain CTO Jack Kendall.
Analog circuits could save power in neural networks in part because they can efficiently perform a key calculation, called multiply and accumulate. That calculation multiplies values from inputs according to various weights, and then it sums all those values up. Two fundamental laws of electrical engineering can basically do that, too. Ohm’s Law multiplies voltage and conductance to give current, and Kirchoff’s Current Law sums the currents entering a point. By storing a neural network’s weights in resistive memory devices, such as memristors, multiply-and-accumulate can happen completely in analog, potentially reducing power consumption by orders of magnitude.
The reason analog AI systems can’t train themselves today has a lot to do with the variability of their components. Just like real neurons, those in analog neural networks don’t all behave exactly alike. To do back propagation with analog components, you must build two separate circuit pathways. One going forward to come up with an answer (called inferencing), the other going backward to do the learning so that the answer becomes more accurate. But because of the variability of analog components, the pathways don't match up.
“You end up accumulating error as you go backwards through the network,” says Bengio. To compensate, a network would need lots of power-hungry analog-to-digital and digital-to-analog circuits, defeating the point of going analog.
Equilibrium propagation allows learning and inferencing to happen on the same network, partly by adjusting the behavior of the network as a whole. “What [equilibrium propagation] allows us to do is to say how we should modify each of these devices so that the overall circuit performs the right thing,” he says. “We turn the physical computation that is happening in the analog devices directly to our advantage.”
Right now, equilibrium propagation is only working in simulation. But Rain plans to have a hardware proof-of-principle in late 2021, according to CEO and cofounder Gordon Wilson. “We are really trying to fundamentally reimagine the hardware computational substrate for artificial intelligence, find the right clues from the brain, and use those to inform the design of this,” he says. The result could be what they call end-to-end analog AI systems that capable of running sophisticated robots or even playing a role in data centers. Both of those are currently considered beyond the capabilities of analog AI, which is now focused only on adding inferencing abilities to sensors and other low-power “edge” devices, while leaving the learning to GPUs. Continue reading →
Posted in Human Robots
#437707 Video Friday: This Robot Will Restock ...
Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here's what we have so far (send us your events!):
CLAWAR 2020 – August 24-26, 2020 – [Online Conference]
ICUAS 2020 – September 1-4, 2020 – Athens, Greece
ICRES 2020 – September 28-29, 2020 – Taipei, Taiwan
AUVSI EXPONENTIAL 2020 – October 5-8, 2020 – [Online Conference]
IROS 2020 – October 25-29, 2020 – Las Vegas, Nev., USA
CYBATHLON 2020 – November 13-14, 2020 – [Online Event]
ICSR 2020 – November 14-16, 2020 – Golden, Colo., USA
Let us know if you have suggestions for next week, and enjoy today's videos.
Tokyo startup Telexistence has recently unveiled a new robot called the Model-T, an advanced teleoperated humanoid that can use tools and grasp a wide range of objects. Japanese convenience store chain FamilyMart plans to test the Model-T to restock shelves in up to 20 stores by 2022. In the trial, a human “pilot” will operate the robot remotely, handling items like beverage bottles, rice balls, sandwiches, and bento boxes.
With Model-T and AWP, FamilyMart and TX aim to realize a completely new store operation by remoteizing and automating the merchandise restocking work, which requires a large number of labor-hours. As a result, stores can operate with less number of workers and enable them to recruit employees regardless of the store’s physical location.
[ Telexistence ]
Quadruped dance-off should be a new robotics competition at IROS or ICRA.
I dunno though, that moonwalk might keep Spot in the lead…
[ Unitree ]
Through a hybrid of simulation and real-life training, this air muscle robot is learning to play table tennis.
Table tennis requires to execute fast and precise motions. To gain precision it is necessary to explore in this high-speed regimes, however, exploration can be safety-critical at the same time. The combination of RL and muscular soft robots allows to close this gap. While robots actuated by pneumatic artificial muscles generate high forces that are required for e.g. smashing, they also offer safe execution of explosive motions due to antagonistic actuation.
To enable practical training without real balls, we introduce Hybrid Sim and Real Training (HYSR) that replays prerecorded real balls in simulation while executing actions on the real system. In this manner, RL can learn the challenging motor control of the PAM-driven robot while executing ~15000 hitting motions.
[ Max Planck Institute ]
Thanks Dieter!
Anthony Cowley wrote in to share his recent thesis work on UPSLAM, a fast and lightweight SLAM technique that records data in panoramic depth images (just PNGs) that are easy to visualize and even easier to share between robots, even on low-bandwidth networks.
[ UPenn ]
Thanks Anthony!
GITAI’s G1 is the space dedicated general-purpose robot. G1 robot will enable automation of various tasks internally & externally on space stations and for lunar base development.
[ Gitai ]
The University of Michigan has a fancy new treadmill that’s built right into the floor, which proves to be a bit much for Mini Cheetah.
But Cassie Blue won’t get stuck on no treadmill! She goes for a 0.3 mile walk across campus, which ends when a certain someone ran the gantry into Cassie Blue’s foot.
[ Michigan Robotics ]
Some serious quadruped research going on at UT Austin Human Centered Robotics Lab.
[ HCRL ]
Will Burrard-Lucas has spent lockdown upgrading his slightly indestructible BeetleCam wildlife photographing robot.
[ Will Burrard-Lucas ]
Teleoperated surgical robots are becoming commonplace in operating rooms, but many are massive (sometimes taking up an entire room) and are difficult to manipulate, especially if a complication arises and the robot needs to removed from the patient. A new collaboration between the Wyss Institute, Harvard University, and Sony Corporation has created the mini-RCM, a surgical robot the size of a tennis ball that weighs as much as a penny, and performed significantly better than manually operated tools in delicate mock-surgical procedures. Importantly, its small size means it is more comparable to the human tissues and structures on which it operates, and it can easily be removed by hand if needed.
[ Harvard Wyss ]
Yaskawa appears to be working on a robot that can scan you with a temperature gun and then jam a mask on your face?
[ Motoman ]
Maybe we should just not have people working in mines anymore, how about that?
[ Exyn ]
Many current human-robot interactive systems tend to use accurate and fast – but also costly – actuators and tracking systems to establish working prototypes that are safe to use and deploy for user studies. This paper presents an embedded framework to build a desktop space for human-robot interaction, using an open-source robot arm, as well as two RGB cameras connected to a Raspberry Pi-based controller that allow a fast yet low-cost object tracking and manipulation in 3D. We show in our evaluations that this facilitates prototyping a number of systems in which user and robot arm can commonly interact with physical objects.
[ Paper ]
IBM Research is proud to host professor Yoshua Bengio — one of the world’s leading experts in AI — in a discussion of how AI can contribute to the fight against COVID-19.
[ IBM Research ]
Ira Pastor, ideaXme life sciences ambassador interviews Professor Dr. Hiroshi Ishiguro, the Director of the Intelligent Robotics Laboratory, of the Department of Systems Innovation, in the Graduate School of Engineering Science, at Osaka University, Japan.
[ ideaXme ]
A CVPR talk from Stanford’s Chelsea Finn on “Generalization in Visuomotor Learning.”
[ Stanford ] Continue reading →
Posted in Human Robots
#437689 GITAI Sending Autonomous Robot to Space ...
We’ve been keeping a close watch on GITAI since early last year—what caught our interest initially is the history of the company, which includes a bunch of folks who started in the JSK Lab at the University of Tokyo, won the DARPA Robotics Challenge Trials as SCHAFT, got swallowed by Google, narrowly avoided being swallowed by SoftBank, and are now designing robots that can work in space.
The GITAI YouTube channel has kept us more to less up to date on their progress so far, and GITAI has recently announced the next step in this effort: The deployment of one of their robots on board the International Space Station in 2021.
Photo: GITAI
GITAI’s S1 is a task-specific 8-degrees-of-freedom arm with an integrated sensing and computing system and 1-meter reach.
GITAI has been working on a variety of robots for space operations, the most sophisticated of which is a humanoid torso called G1, which is controlled through an immersive telepresence system. What will be launching into space next year is a more task-specific system called the S1, which is an 8-degrees-of-freedom arm with an integrated sensing and computing system that can be wall-mounted and has a 1-meter reach.
The S1 will be living on board a commercially funded, pressurized airlock-extension module called Bishop, developed by NanoRacks. Mounted on the inside of the Bishop module, the S1 will have access to a task board and a small assembly area, where it will demonstrate common crew intra-vehicular activity, or IVA—tasks like flipping switches, turning knobs, and managing cables. It’ll also do some in-space assembly, or ISA, attaching panels to create a solar array.
Here’s a demonstration of some task board activities, conducted on Earth in a mockup of Bishop:
GITAI says that “all operations conducted by the S1 GITAI robotic arm will be autonomous, followed by some teleoperations from Nanoracks’ in-house mission control.” This is interesting, because from what we’ve seen until now, GITAI has had a heavy emphasis on telepresence, with a human in the loop to get stuff done. As GITAI’s founder and CEO Sho Nakanose commented to us a year ago, “Telepresence robots have far better performance and can be made practical much quicker than autonomous robots, so first we are working on making telepresence robots practical.”
So what’s changed? “GITAI has been concentrating on teleoperations to demonstrate the dexterity of our robot, but now it’s time to show our capabilities to do the same this time with autonomy,” Nakanose told us last week. “In an environment with minimum communication latency, it would be preferable to operate a robot more with teleoperations to enhance the capability of the robot, since with the current technology level of AI, what a robot can do autonomously is very limited. However, in an environment where the latency becomes noticeable, it would become more efficient to have a mixture of autonomy and teleoperations depending on the application. Eventually, in an ideal world, a robot will operate almost fully autonomously with minimum human cognizance.”
“In an environment where the latency becomes noticeable, it would become more efficient to have a mixture of autonomy and teleoperations depending on the application. Eventually, in an ideal world, a robot will operate almost fully autonomously with minimum human cognizance.”
—Sho Nakanose, GITAI founder and CEO
Nakanose says that this mission will help GITAI to “acquire the skills, know-how, and experience necessary to prepare a robot to be ISS compatible, prov[ing] the maturity of our technology in the microgravity environment.” Success would mean conducting both IVA and ISA experiments as planned (autonomous and teleop for IVA, fully autonomous for ISA), which would be pretty awesome, but we’re told that GITAI has already received a research and development order for space robots from a private space company, and Nakanose expects that “by the mid-2020s, we will be able to show GITAI's robots working in space on an actual mission.”
NanoRacks is schedule to launch the Bishop module on SpaceX CRS-21 in November. The S1 will be launched separately in 2021, and a NASA astronaut will install the robot and then leave it alone to let it start demonstrating how work in space can be made both safer and cheaper once the humans have gotten out of the way. Continue reading →
Posted in Human Robots
#437671 Video Friday: Researchers 3D Print ...
Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here’s what we have so far (send us your events!):
ICRES 2020 – September 28-29, 2020 – Taipei, Taiwan
AUVSI EXPONENTIAL 2020 – October 5-8, 2020 – [Online]
IROS 2020 – October 25-29, 2020 – [Online]
ROS World 2020 – November 12, 2020 – [Online]
CYBATHLON 2020 – November 13-14, 2020 – [Online]
ICSR 2020 – November 14-16, 2020 – Golden, Colo., USA
Let us know if you have suggestions for next week, and enjoy today’s videos.
The Giant Gundam in Yokohama is actually way cooler than I thought it was going to be.
[ Gundam Factory ] via [ YouTube ]
A new 3D-printing method will make it easier to manufacture and control the shape of soft robots, artificial muscles and wearable devices. Researchers at UC San Diego show that by controlling the printing temperature of liquid crystal elastomer, or LCE, they can control the material’s degree of stiffness and ability to contract—also known as degree of actuation. What’s more, they are able to change the stiffness of different areas in the same material by exposing it to heat.
[ UCSD ]
Thanks Ioana!
This is the first successful reactive stepping test on our new torque-controlled biped robot named Bolt. The robot has 3 active degrees of freedom per leg and one passive joint in ankle. Since there is no active joint in ankle, the robot only relies on step location and timing adaptation to stabilize its motion. Not only can the robot perform stepping without active ankles, but it is also capable of rejecting external disturbances as we showed in this video.
[ ODRI ]
The curling robot “Curly” is the first AI-based robot to demonstrate competitive curling skills in an icy real environment with its high uncertainties. Scientists from seven different Korean research institutions including Prof. Klaus-Robert Müller, head of the machine-learning group at TU Berlin and guest professor at Korea University, have developed an AI-based curling robot.
[ TU Berlin ]
MoonRanger, a small robotic rover being developed by Carnegie Mellon University and its spinoff Astrobotic, has completed its preliminary design review in preparation for a 2022 mission to search for signs of water at the moon’s south pole. Red Whittaker explains why the new MoonRanger Lunar Explorer design is innovative and different from prior planetary rovers.
[ CMU ]
Cobalt’s security robot can now navigate unmodified elevators, which is an impressive feat.
Also, EXTERMINATE!
[ Cobalt ]
OrionStar, the robotics company invested in by Cheetah Mobile, announced the Robotic Coffee Master. Incorporating 3,000 hours of AI learning, 30,000 hours of robotic arm testing and machine vision training, the Robotic Coffee Master can perform complex brewing techniques, such as curves and spirals, with millimeter-level stability and accuracy (reset error ≤ 0.1mm).
[ Cheetah Mobile ]
DARPA OFFensive Swarm-Enabled Tactics (OFFSET) researchers recently tested swarms of autonomous air and ground vehicles at the Leschi Town Combined Arms Collective Training Facility (CACTF), located at Joint Base Lewis-McChord (JBLM) in Washington. The Leschi Town field experiment is the fourth of six planned experiments for the OFFSET program, which seeks to develop large-scale teams of collaborative autonomous systems capable of supporting ground forces operating in urban environments.
[ DARPA ]
Here are some highlights from Team Explorer’s SubT Urban competition back in February.
[ Team Explorer ]
Researchers with the Skoltech Intelligent Space Robotics Laboratory have developed a system that allows easy interaction with a micro-quadcopter with LEDs that can be used for light-painting. The researchers used a 92x92x29 mm Crazyflie 2.0 quadrotor that weighs just 27 grams, equipped with a light reflector and an array of controllable RGB LEDs. The control system consists of a glove equipped with an inertial measurement unit (IMU; an electronic device that tracks the movement of a user’s hand), and a base station that runs a machine learning algorithm.
[ Skoltech ]
“DeKonBot” is the prototype of a cleaning and disinfection robot for potentially contaminated surfaces in buildings such as door handles, light switches or elevator buttons. While other cleaning robots often spray the cleaning agents over a large area, DeKonBot autonomously identifies the surface to be cleaned.
[ Fraunhofer IPA ]
On Oct. 20, the OSIRIS-REx mission will perform the first attempt of its Touch-And-Go (TAG) sample collection event. Not only will the spacecraft navigate to the surface using innovative navigation techniques, but it could also collect the largest sample since the Apollo missions.
[ NASA ]
With all the robotics research that seems to happen in places where snow is more of an occasional novelty or annoyance, it’s good to see NORLAB taking things more seriously
[ NORLAB ]
Telexistence’s Model-T robot works very slowly, but very safely, restocking shelves.
[ Telexistence ] via [ YouTube ]
Roboy 3.0 will be unveiled next month!
[ Roboy ]
KUKA ready2_educate is your training cell for hands-on education in robotics. It is especially aimed at schools, universities and company training facilities. The training cell is a complete starter package and your perfect partner for entry into robotics.
[ KUKA ]
A UPenn GRASP Lab Special Seminar on Data Driven Perception for Autonomy, presented by Dapo Afolabi from UC Berkeley.
Perception systems form a crucial part of autonomous and artificial intelligence systems since they convert data about the relationship between an autonomous system and its environment into meaningful information. Perception systems can be difficult to build since they may involve modeling complex physical systems or other autonomous agents. In such scenarios, data driven models may be used to augment physics based models for perception. In this talk, I will present work making use of data driven models for perception tasks, highlighting the benefit of such approaches for autonomous systems.
[ GRASP Lab ]
A Maryland Robotics Center Special Robotics Seminar on Underwater Autonomy, presented by Ioannis Rekleitis from the University of South Carolina.
This talk presents an overview of algorithmic problems related to marine robotics, with a particular focus on increasing the autonomy of robotic systems in challenging environments. I will talk about vision-based state estimation and mapping of underwater caves. An application of monitoring coral reefs is going to be discussed. I will also talk about several vehicles used at the University of South Carolina such as drifters, underwater, and surface vehicles. In addition, a short overview of the current projects will be discussed. The work that I will present has a strong algorithmic flavour, while it is validated in real hardware. Experimental results from several testing campaigns will be presented.
[ MRC ]
This week’s CMU RI Seminar comes from Scott Niekum at UT Austin, on Scaling Probabilistically Safe Learning to Robotics.
Before learning robots can be deployed in the real world, it is critical that probabilistic guarantees can be made about the safety and performance of such systems. This talk focuses on new developments in three key areas for scaling safe learning to robotics: (1) a theory of safe imitation learning; (2) scalable reward inference in the absence of models; (3) efficient off-policy policy evaluation. The proposed algorithms offer a blend of safety and practicality, making a significant step towards safe robot learning with modest amounts of real-world data.
[ CMU RI ] Continue reading →
Posted in Human Robots