Tag Archives: scan

#439095 DARPA Prepares for the Subterranean ...

The DARPA Subterranean Challenge Final Event is scheduled to take place at the Louisville Mega Cavern in Louisville, Kentucky, from September 21 to 23. We’ve followed SubT teams as they’ve explored their way through abandoned mines, unfinished nuclear reactors, and a variety of caves, and now everything comes together in one final course where the winner of the Systems Track will take home the $2 million first prize.

It’s a fitting reward for teams that have been solving some of the hardest problems in robotics, but winning isn’t going to be easy, and we’ll talk with SubT Program Manager Tim Chung about what we have to look forward to.

Since we haven’t talked about SubT in a little while (what with the unfortunate covid-related cancellation of the Systems Track Cave Circuit), here’s a quick refresher of where we are: the teams have made it through the Tunnel Circuit, the Urban Circuit, and a virtual version of the Cave Circuit, and some of them have been testing in caves of their own. The Final Event will include all of these environments, and the teams of robots will have 60 minutes to autonomously map the course, locating artifacts to score points. Since I’m not sure where on Earth there’s an underground location that combines tunnels and caves with urban structures, DARPA is going to have to get creative, and the location in which they’ve chosen to do that is Louisville, Kentucky.

The Louisville Mega Cavern is a former limestone mine, most of which is under the Louisville Zoo. It’s not all that deep, mostly less than 30 meters under the surface, but it’s enormous: with 370,000 square meters of rooms and passages, the cavern currently hosts (among other things) a business park, a zipline course, and mountain bike trails, because why not. While DARPA is keeping pretty quiet on the details, I’m guessing that they’ll be taking over a chunk of the cavern and filling it with features representing as many of the environmental challenges as they can.

To learn more about how the SubT Final Event is going to go, we spoke with SubT Program Manager Tim Chung. But first, we talked about Tim’s perspective on the success of the Urban Circuit, and how teams have been managing without an in-person Cave Circuit.

IEEE Spectrum: How did the SubT Urban Circuit go?

Tim Chung: On a couple fronts, Urban Circuit was really exciting. We were in this unfinished nuclear power plant—I’d be surprised if any of the competitors had prior experience in such a facility, or anything like it. I think that was illuminating both from an experiential point of view for the competitors, but also from a technology point of view, too.

One thing that I thought was really interesting was that we, DARPA, didn't need to make the venue more challenging. The real world is really that hard. There are places that were just really heinous for these robots to have to navigate through in order to look in every nook and cranny for artifacts. There were corners and doorways and small corridors and all these kind of things that really forced the teams to have to work hard, and the feedback was, why did DARPA have to make it so hard? But we didn’t, and in fact there were places that for the safety of the robots and personnel, we had to ensure the robots couldn’t go.

It sounds like some teams thought this course was on the more difficult side—do you think you tuned it to just the right amount of DARPA-hard?

Our calibration worked quite well. We were able to tease out and help refine and better understand what technologies are both useful and critical and also those technologies that might not necessarily get you the leap ahead capability. So as an example, the Urban Circuit really emphasized verticality, where you have to be able to sense, understand, and maneuver in three dimensions. Being able to capitalize on their robot technologies to address that verticality really stratified the teams, and showed how critical those capabilities are.

We saw teams that brought a lot of those capabilities do very well, and teams that brought baseline capabilities do what they could on the single floor that they were able to operate on. And so I think we got the Goldilocks solution for Urban Circuit that combined both difficulty and ambition.

Photos: Evan Ackerman/IEEE Spectrum

Two SubT Teams embedded networking equipment in balls that they could throw onto the course.

One of the things that I found interesting was that two teams independently came up with throwable network nodes. What was DARPA’s reaction to this? Is any solution a good solution, or was it more like the teams were trying to game the system?

You mean, do we want teams to game the rules in any way so as to get a competitive advantage? I don't think that's what the teams were doing. I think they were operating not only within the bounds of the rules, which permitted such a thing as throwable sensors where you could stand at the line and see how far you could chuck these things—not only was that acceptable by the rules, but anticipated. Behind the scenes, we tried to do exactly what these teams are doing and think through different approaches, so we explicitly didn't forbid such things in our rules because we thought it's important to have as wide an aperture as possible.

With these comms nodes specifically, I think they’re pretty clever. They were in some cases hacked together with a variety of different sports paraphernalia to see what would provide the best cushioning. You know, a lot of that happens in the field, and what it captured was that sometimes you just need to be up at two in the morning and thinking about things in a slightly different way, and that's when some nuggets of innovation can arise, and we see this all the time with operators in the field as well. They might only have duct tape or Styrofoam or whatever the case may be and that's when they come up with different ways to solve these problems. I think from DARPA’s perspective, and certainly from my perspective, wherever innovation can strike, we want to try to encourage and inspire those opportunities. I thought it was great, and it’s all part of the challenge.

Is there anything you can tell us about what your original plan had been for the Cave Circuit?

I can say that we’ve had the opportunity to go through a number of these caves scattered all throughout the country, and engage with caving communities—cavers clubs, speleologists that conduct research, and then of course the cave rescue community. The single biggest takeaway
is that every cave, and there are tens of thousands of them in the US alone, every cave has its own personality, and a lot of that personality is quite hidden from humans, because we can’t explore or access all of the cave. This led us to a number of different caves that were intriguing from a DARPA perspective but also inspirational for our Cave Circuit Virtual Competition.

How do you feel like the tuning was for the Virtual Cave Circuit?

The Virtual Competition, as you well know, was exciting in the sense that we could basically combine eight worlds into one competition, whereas the systems track competition really didn’t give us that opportunity. Even if we were able have held the Cave Circuit Systems Competition in person, it would have been at one site, and it would have been challenging to represent the level of diversity that we could with the Virtual Competition. So I think from that perspective, it’s clearly an advantage in terms of calibration—diversity gets you the ability to aggregate results to capture those that excel across all worlds as well as those that do well in one world or some worlds and not the others. I think the calibration was great in the sense that we were able to see the gamut of performance. Those that did well, did quite well, and those that have room to grow showed where those opportunities are for them as well.

We had to find ways to capture that diversity and that representativeness, and I think one of the fun ways we did that was with the different cave world tiles that we were able to combine in a variety of different ways. We also made use of a real world data set that we were able to take from a laser scan. Across the board, we had a really great chance to illustrate why virtual testing and simulation still plays such a dominant role in robotics technology development, and why I think it will continue to play an increasing role for developing these types of autonomy solutions.

Photo: Team CSIRO Data 61

How can systems track teams learn from their testing in whatever cave is local to them and effectively apply that to whatever cave environment is part of the final considering what the diversity of caves is?

I think that hits the nail on the head for what we as technologists are trying to discover—what are the transferable generalizable insights and how does that inform our technology development? As roboticists we want to optimize our systems to perform well at the tasks that they were designed to do, and oftentimes that means specialization because we get increased performance at the expense of being a generalist robot. I think in the case of SubT, we want to have our cake and eat it too—we want robots that perform well and reliably, but we want them to do so not just in one environment, which is how we tend to think about robot performance, but we want them to operate well in many environments, many of which have yet to be faced.

And I think that's kind of the nuance here, that we want robot systems to be generalists for the sake of being able to handle the unknown, namely the real world, but still achieve a high level of performance and perhaps they do that to their combined use of different technologies or advances in autonomy or perception approaches or novel mechanisms or mobility, but somehow they're still able, at least in aggregate, to achieve high performance.

We know these teams eagerly await any type of clue that DARPA can provide like about the SubT environments. From the environment previews for Tunnel, Urban, and even Cave, the teams were pivoting around and thinking a little bit differently. The takeaway, however, was that they didn't go to a clean sheet design—their systems were flexible enough that they could incorporate some of those specialist trends while still maintaining the notion of a generalist framework.

Looking ahead to the SubT Final, what can you tell us about the Louisville Mega Cavern?

As always, I’ll keep you in suspense until we get you there, but I can say that from the beginning of the SubT Challenge we had always envisioned teams of robots that are able to address not only the uncertainty of what's right in front of them, but also the uncertainty of what comes next. So I think the teams will be advantaged by thinking through subdomain awareness, or domain awareness if you want to generalize it, whether that means tuning multi-purpose robots, or deploying different robots, or employing your team of robots differently. Knowing which subdomain you are in is likely to be helpful, because then you can take advantage of those unique lessons learned through all those previous experiences then capitalize on that.

As far as specifics, I think the Mega Cavern offers many of the features important to what it means to be underground, while giving DARPA a pretty blank canvas to realize our vision of the SubT Challenge.

The SubT Final will be different from the earlier circuits in that there’s just one 60-minute run, rather than two. This is going to make things a lot more stressful for teams who have experienced bad robot days—why do it this way?

The preliminary round has two 30-minute runs, and those two runs are very similar to how we have done it during the circuits, of a single run per configuration per course. Teams will have the opportunity to show that their systems can face the obstacles in the final course, and it's the sum of those scores much like we did during the circuits, to help mitigate some of the concerns that you mentioned of having one robot somehow ruin their chances at a prize.

The prize round does give DARPA as well as the community a chance to focus on the top six teams from the preliminary round, and allows us to understand how they came to be at the top of the pack while emphasizing their technological contributions. The prize round will be one and done, but all of these teams we anticipate will be putting their best robot forward and will show the world why they deserve to win the SubT Challenge.

We’ve always thought that when called upon these robots need to operate in really challenging environments, and in the context of real world operations, there is no second chance. I don't think it's actually that much of a departure from our interests and insistence on bringing reliable technologies to the field, and those teams that might have something break here and there, that's all part of the challenge, of being resilient. Many teams struggled with robots that were debilitated on the course, and they still found ways to succeed and overcome that in the field, so maybe the rules emphasize that desire for showing up and working on game day which is consistent, I think, with how we've always envisioned it. This isn’t to say that these systems have to work perfectly, they just have to work in a way such that the team is resilient enough to tackle anything that they face.

It’s not too late for teams to enter for both the Virtual Track and the Systems Track to compete in the SubT Final, right?

Yes, that's absolutely right. Qualifications are still open, we are eager to welcome new teams to join in along with our existing competitors. I think any dark horse competitors coming into the Finals may be able to bring something that we haven't seen before, and that would be really exciting. I think it'll really make for an incredibly vibrant and illuminating final event.

The final event qualification deadline for the Systems Competition is April 21, and the qualification deadline for the Virtual Competition is June 29. More details here. Continue reading

Posted in Human Robots

#437924 How a Software Map of the Entire Planet ...

i
“3D map data is the scaffolding of the 21st century.”

–Edward Miller, Founder, Scape Technologies, UK

Covered in cameras, sensors, and a distinctly spaceship looking laser system, Google’s autonomous vehicles were easy to spot when they first hit public roads in 2015. The key hardware ingredient is a spinning laser fixed to the roof, called lidar, which provides the car with a pair of eyes to see the world. Lidar works by sending out beams of light and measuring the time it takes to bounce off objects back to the source. By timing the light’s journey, these depth-sensing systems construct fully 3D maps of their surroundings.

3D maps like these are essentially software copies of the real world. They will be crucial to the development of a wide range of emerging technologies including autonomous driving, drone delivery, robotics, and a fast-approaching future filled with augmented reality.

Like other rapidly improving technologies, lidar is moving quickly through its development cycle. What was an expensive technology on the roof of a well-funded research project is now becoming cheaper, more capable, and readily available to consumers. At some point, lidar will come standard on most mobile devices and is now available to early-adopting owners of the iPhone 12 Pro.

Consumer lidar represents the inevitable shift from wealthy tech companies generating our world’s map data, to a more scalable crowd-sourced approach. To develop the repository for their Street View Maps product, Google reportedly spent $1-2 billion sending cars across continents photographing every street. Compare that to a live-mapping service like Waze, which uses crowd-sourced user data from its millions of users to generate accurate and real-time traffic conditions. Though these maps serve different functions, one is a static, expensive, unchanging map of the world while the other is dynamic, real-time, and constructed by users themselves.

Soon millions of people may be scanning everything from bedrooms to neighborhoods, resulting in 3D maps of significant quality. An online search for lidar room scans demonstrates just how richly textured these three-dimensional maps are compared to anything we’ve had before. With lidar and other depth-sensing systems, we now have the tools to create exact software copies of everywhere and everything on earth.

At some point, likely aided by crowdsourcing initiatives, these maps will become living breathing, real-time representations of the world. Some refer to this idea as a “digital twin” of the planet. In a feature cover story, Kevin Kelly, the cofounder of Wired magazine, calls this concept the “mirrorworld,” a one-to-one software map of everything.

So why is that such a big deal? Take augmented reality as an example.

Of all the emerging industries dependent on such a map, none are more invested in seeing this concept emerge than those within the AR landscape. Apple, for example, is not-so-secretly developing a pair of AR glasses, which they hope will deliver a mainstream turning point for the technology.

For Apple’s AR devices to work as anticipated, they will require virtual maps of the world, a concept AR insiders call the “AR cloud,” which is synonymous with the “mirrorworld” concept. These maps will be two things. First, they will be a tool that creators use to place AR content in very specific locations; like a world canvas to paint on. Second, they will help AR devices both locate and understand the world around them so they can render content in a believable way.

Imagine walking down a street wanting to check the trading hours of a local business. Instead of pulling out your phone to do a tedious search online, you conduct the equivalent of a visual google search simply by gazing at the store. Albeit a trivial example, the AR cloud represents an entirely non-trivial new way of managing how we organize the world’s information. Access to knowledge can be shifted away from the faraway monitors in our pocket, to its relevant real-world location.

Ultimately this describes a blurring of physical and digital infrastructure. Our public and private spaces will thus be comprised equally of both.

No example demonstrates this idea better than Pokémon Go. The game is straightforward enough; users capture virtual characters scattered around the real world. Today, the game relies on traditional GPS technology to place its characters, but GPS is accurate only to within a few meters of a location. For a car navigating on a highway or locating Pikachus in the world, that level of precision is sufficient. For drone deliveries, driverless cars, or placing a Pikachu in a specific location, say on a tree branch in a park, GPS isn’t accurate enough. As astonishing as it may seem, many experimental AR cloud concepts, even entirely mapped cities, are location specific down to the centimeter.

Niantic, the $4 billion publisher behind Pokémon Go, is aggressively working on developing a crowd-sourced approach to building better AR Cloud maps by encouraging their users to scan the world for them. Their recent acquisition of 6D.ai, a mapping software company developed by the University of Oxford’s Victor Prisacariu through his work at Oxford’s Active Vision Lab, indicates Niantic’s ambition to compete with the tech giants in this space.

With 6D.ai’s technology, Niantic is developing the in-house ability to generate their own 3D maps while gaining better semantic understanding of the world. By going beyond just knowing there’s a temporary collection of orange cones in a certain location, for example, the game may one day understand the meaning behind this; that a temporary construction zone means no Pokémon should spawn here to avoid drawing players to this location.

Niantic is not the only company working on this. Many of the big tech firms you would expect have entire teams focused on map data. Facebook, for example, recently acquired the UK-based Scape technologies, a computer vision startup mapping entire cities with centimeter precision.

As our digital maps of the world improve, expect a relentless and justified discussion of privacy concerns as well. How will society react to the idea of a real-time 3D map of their bedroom living on a Facebook or Amazon server? Those horrified by the use of facial recognition AI being used in public spaces are unlikely to find comfort in the idea of a machine-readable world subject to infinite monitoring.

The ability to build high-precision maps of the world could reshape the way we engage with our planet and promises to be one of the biggest technology developments of the next decade. While these maps may stay hidden as behind-the-scenes infrastructure powering much flashier technologies that capture the world’s attention, they will soon prop up large portions of our technological future.

Keep that in mind when a car with no driver is sharing your road.

Image credit: sergio souza / Pexels Continue reading

Posted in Human Robots

#437851 Boston Dynamics’ Spot Robot Dog ...

Boston Dynamics has been fielding questions about when its robots are going to go on sale and how much they’ll cost for at least a dozen years now. I can say this with confidence, because that’s how long I’ve been a robotics journalist, and I’ve been pestering them about it the entire time. But it’s only relatively recently that the company started to make a concerted push away from developing robots exclusively for the likes of DARPA into platforms with more commercial potential, starting with a compact legged robot called Spot, first introduced in 2016.

Since then, we’ve been following closely as Spot has gone from a research platform to a product, and today, Boston Dynamics is announcing the final step in that process: commercial availability. You can now order a Spot Explorer Kit from the Boston Dynamics online store for US $74,500 (plus tax), shipping included, with delivery in 6 to 8 weeks. FINALLY!

Over the past 10 months or so, Boston Dynamics has leased Spot robots to carefully selected companies, research groups, and even a few individuals as part of their early adopter program—that’s where all of the clips in the video below came from. While there are over 100 Spots out in the world right now, getting one of them has required convincing Boston Dynamics up front that you knew more or less exactly what you wanted to do and how you wanted to do it. If you’re a big construction company or the Jet Propulsion Laboratory or Adam Savage, that’s all well and good, but for other folks who think that a Spot could be useful for them somehow and want to give it a shot, this new availability provides a fewer-strings attached opportunity to do some experimentation with the robot.

There’s a lot of cool stuff going on in that video, but we were told that the one thing that really stood out to the folks at Boston Dynamics was a 2-second clip that you can see on the left-hand side of the screen from 0:19 to 0:21. In it, Spot is somehow managing to walk across a spider web of rebar without getting tripped up, at faster than human speed. This isn’t something that Spot was specifically programmed to do, and in fact the Spot User Guide specifically identifies “rebar mesh” as an unsafe operating environment. But the robot just handles it, and that’s a big part of what makes Spot so useful—its ability to deal with (almost) whatever you can throw at it.

Before you get too excited, Boston Dynamics is fairly explicit that the current license for the robot is intended for commercial use, and the company specifically doesn’t want people to be just using it at home for fun. We know this because we asked (of course we asked), and they told us “we specifically don’t want people to just be using it at home for fun.” Drat. You can still buy one as an individual, but you have to promise that you’ll follow the terms of use and user guidelines, and it sounds like using a robot in your house might be the second-fastest way to invalidate your warranty:

SPOT IS AN AMAZING ROBOT, BUT IS NOT CERTIFIED SAFE FOR IN-HOME USE OR INTENDED FOR USE NEAR CHILDREN OR OTHERS WHO MAY NOT APPRECIATE THE HAZARDS ASSOCIATED WITH ITS OPERATION.

Not being able to get Spot to play with your kids may be disappointing, but for those of you with the sort of kids who are also students, the good news is that Boston Dynamics has carved out a niche for academic institutions, which can buy Spot at a discounted price. And if you want to buy a whole pack of Spots, there’s a bulk discount for Enterprise users as well.

What do you get for $74,500? All this!

Spot robot
Spot battery (2x)
Spot charger
Tablet controller and charger
Robot case for storage and transportation
FREE SHIPPING!

Photo: Boston Dynamics

The basic package includes the robot, two batteries, charger, a tablet controller, and a storage case.

You can view detailed specs here.

So is $75k a lot of money for a robot like Spot, or not all that much? We don’t have many useful points of comparison, partially because it’s not clear to what extent other pre-commercial quadrupedal robots (like ANYmal or Aliengo) share capabilities and features with Spot. For more perspective on Spot’s price tag, we spoke to Michael Perry, vice president of business development at Boston Dynamics.

IEEE Spectrum: Why is Spot so affordable?

Michael Perry: The main goal of selling the robot at this stage is to try to get it into the hands of as many application developers as possible, so that we can learn from the community what the biggest driver of value is for Spot. As a platform, unlocking the value of an ecosystem is our core focus right now.

Spectrum: Why is Spot so expensive?

Perry: Expensive is relative, but compared to the initial prototypes of Spot, we’ve been able to drop down the cost pretty significantly. One key thing has been designing it for robustness—we’ve put hundreds and hundreds of hours on the robot to make sure that it’s able to be successful when it falls, or when it has an electrostatic discharge. We’ve made sure that it’s able to perceive a wide variety of environments that are difficult for traditional vision-based sensors to handle. A lot of that engineering is baked into the core product so that you don’t have to worry about the mobility or robotic side of the equation, you can just focus on application development.

Photos: Boston Dynamics

Accessories for Spot include [clockwise from top left]: Spot GXP with additional ports for payload integration; Spot CAM with panorama camera and advanced comms; Spot CAM+ with pan-tilt-zoom camera for inspections; Spot EAP with lidar to enhance autonomy on large sites; Spot EAP+ with Spot CAM camera plus lidar; and Spot CORE for additional processing power.

The $75k that you’ll pay for the Spot Explorer Kit, it’s important to note, is just the base price for the robot. As with other things that fall into this price range (like a luxury car), there are all kinds of fun ways to drive that cost up with accessories, although for Spot, some of those accessories will be necessary for many (if not most) applications. For example, a couple of expansion ports to make it easier to install your own payloads on Spot will run you $1,275. An additional battery is $4,620. And if you want to really get some work done, the Enhanced Autonomy Package (with 360 cameras, lights, better comms, and a Velodyne VLP-16) will set you back an additional $34,570. If you were hoping for an arm, you’ll have to wait until the end of the year.

Each Spot also includes a year’s worth of software updates and a warranty, although the standard warranty just covers “defects related to materials and workmanship” not “I drove my robot off a cliff” or “I tried to take my robot swimming.” For that sort of thing (user error) to be covered, you’ll need to upgrade to the $12,000 Spot CARE premium service plan to cover your robot for a year as long as you don’t subject it to willful abuse, which both of those examples I just gave probably qualify as.

While we’re on the subject of robot abuse, Boston Dynamics has very sensibly devoted a substantial amount of the Spot User Guide to help new users understand how they should not be using their robot, in order to “lessen the risk of serious injury, death, or robot and other property damage.” According to the guide, some things that could cause Spot to fall include holes, cliffs, slippery surfaces (like ice and wet grass), and cords. Spot’s sensors also get confused by “transparent, mirrored, or very bright obstacles,” and the guide specifically says Spot “may crash into glass doors and windows.” Also this: “Spot cannot predict trajectories of moving objects. Do not operate Spot around moving objects such as vehicles, children, or pets.”

We should emphasize that this is all totally reasonable, and while there are certainly a lot of things to be aware of, it’s frankly astonishing that these are the only things that Boston Dynamics explicitly warns users against. Obviously, not every potentially unsafe situation or thing is described above, but the point is that Boston Dynamics is willing to say to new users, “here’s your robot, go do stuff with it” without feeling the need to hold their hand the entire time.

There’s one more thing to be aware of before you decide to buy a Spot, which is the following:

“All orders will be subject to Boston Dynamics’ Terms and Conditions of Sale which require the beneficial use of its robots.”

Specifically, this appears to mean that you aren’t allowed to (or supposed to) use the robot in a way that could hurt living things, or “as a weapon, or to enable any weapon.” The conditions of sale also prohibit using the robot for “any illegal or ultra-hazardous purpose,” and there’s some stuff in there about it not being cool to use Spot for “nuclear, chemical, or biological weapons proliferation, or development of missile technology,” which seems weirdly specific.

“Once you make a technology more broadly available, the story of it starts slipping out of your hands. Our hope is that ahead of time we’re able to clearly articulate the beneficial uses of the robot in environments where we think the robot has a high potential to reduce the risk to people, rather than potentially causing harm.”
—Michael Perry, Boston Dynamics

I’m very glad that Boston Dynamics is being so upfront about requiring that Spot is used beneficially. However, it does put the company in a somewhat challenging position now that these robots are being sold. Boston Dynamics can (and will) perform some amount of due-diligence before shipping a Spot, but ultimately, once the robots are in someone else’s hands, there’s only so much that BD can do.

Spectrum: Why is beneficial use important to Boston Dynamics?

Perry: One of the key things that we’ve highlighted many times in our license and terms of use is that we don’t want to see the robot being used in any way that inflicts physical harm on people or animals. There are philosophical reasons for that—I think all of us don’t want to see our technology used in a way that would hurt people. But also from a business perspective, robots are really terrible at conveying intention. In order for the robot to be helpful long-term, it has to be trusted as a piece of technology. So rather than looking at a robot and wondering, “is this something that could potentially hurt me,” we want people to think “this is a robot that’s here to help me.” To the extent that people associate Boston Dynamics with cutting edge robots, we think that this is an important stance for the rollout of our first commercial product. If we find out that somebody’s violated our terms of use, their warranty is invalidated, we won’t repair their product, and we have a licensing timeout that would prevent them from accessing their robot after that timeout has expired. It’s a remediation path, but we do think that it’s important to at least provide that as something that helps enforce our position on use of our technology.

It’s very important to keep all of this in context: Spot is a tool. It’s got some autonomy and the appearance of agency, but it’s still just doing what people tell it to do, even if those things might be unsafe. If you read through the user guide, it’s clear how much of an effort Boston Dynamics is making to try to convey the importance of safety to Spot users—and ultimately, barring some unforeseen and catastrophic software or hardware issues, safety is about the users, rather than Boston Dynamics or Spot itself. I bring this up because as we start seeing more and more Spots doing things without Boston Dynamics watching over them quite so closely, accidents are likely inevitable. Spot might step on someone’s foot. It might knock someone over. If Spot was perfectly safe, it wouldn’t be useful, and we have to acknowledge that its impressive capabilities come with some risks, too.

Photo: Boston Dynamics

Each Spot includes a year’s worth of software updates and a warranty, although the standard warranty just covers “defects related to materials and workmanship” not “I drove my robot off a cliff.”

Now that Spot is on the market for real, we’re excited to see who steps up and orders one. Depending on who the potential customer is, Spot could either seem like an impossibly sophisticated piece of technology that they’d never be able to use, or a magical way of solving all of their problems overnight. In reality, it’s of course neither of those things. For the former (folks with an idea but without a lot of robotics knowledge or experience), Spot does a lot out of the box, but BD is happy to talk with people and facilitate connections with partners who might be able to integrate specific software and hardware to get Spot to do a unique task. And for the latter (who may also be folks with an idea but without a lot of robotics knowledge or experience), BD’s Perry offers a reminder Spot is not Rosie the Robot, and would be equally happy to talk about what the technology is actually capable of doing.

Looking forward a bit, we asked Perry whether Spot’s capabilities mean that customers are starting to think beyond using robots to simply replace humans, and are instead looking at them as a way of enabling a completely different way of getting things done.

Spectrum: Do customers interested in Spot tend to think of it as a way of replacing humans at a specific task, or as a system that can do things that humans aren’t able to do?

Perry: There are what I imagine as three levels of people understanding the robot applications. Right now, we’re at level one, where you take a person out of this dangerous, dull job, and put a robot in. That’s the entry point. The second level is, using the robot, can we increase the production of that task? For example, take site documentation on a construction site—right now, people do 360 image capture of a site maybe once a week, and they might do a laser scan of the site once per project. At the second level, the question is, what if you were able to get that data collection every day, or multiple times a day? What kinds of benefits would that add to your process? To continue the construction example, the third level would be, how could we completely redesign this space now that we know that this type of automation is available? To take one example, there are some things that we cannot physically build because it’s too unsafe for people to be a part of that process, but if you were to apply robotics to that process, then you could potentially open up a huge envelope of design that has been inaccessible to people.

To order a Spot of your very own, visit shop.bostondynamics.com.

A version of this post appears in the August 2020 print issue as “$74,500 Will Fetch You a Spot.” Continue reading

Posted in Human Robots

#437783 Ex-Googler’s Startup Comes Out of ...

Over the last 10 years, the PR2 has helped roboticists make an enormous amount of progress in mobile manipulation over a relatively short time. I mean, it’s been a decade already, but still—robots are hard, and giving a bunch of smart people access to a capable platform where they didn’t have to worry about hardware and could instead focus on doing interesting and useful things helped to establish a precedent for robotics research going forward.

Unfortunately, not everyone can afford an enormous US $400,000 robot, and even if they could, PR2s are getting very close to the end of their lives. There are other mobile manipulators out there taking the place of the PR2, but so far, size and cost have largely restricted them to research labs. Lots of good research is being done, but it’s getting to the point where folks want to take the next step: making mobile manipulators real-world useful.

Today, a company called Hello Robot is announcing a new mobile manipulator called the Stretch RE1. With offices in the San Francisco Bay Area and in Atlanta, Ga., Hello Robot is led by Aaron Edsinger and Charlie Kemp, and by combining decades of experience in industry and academia they’ve managed to come up with a robot that’s small, lightweight, capable, and affordable, all at the same time. For now, it’s a research platform, but eventually, its creators hope that it will be able to come into our homes and take care of us when we need it to.

A fresh look at mobile manipulators
To understand the concept behind Stretch, it’s worth taking a brief look back at what Edsinger and Kemp have been up to for the past 10 years. Edsinger co-founded Meka Robotics in 2007, which built expensive, high performance humanoid arms, torsos, and heads for the research market. Meka was notable for being the first robotics company (as far as we know) to sell robot arms that used series elastic actuators, and the company worked extensively with Georgia Tech researchers. In 2011, Edsinger was one of the co-founders of Redwood Robotics (along with folks from SRI and Willow Garage), which was going to develop some kind of secret and amazing new robot arm before Google swallowed it in late 2013. At the same time, Google also acquired Meka and a bunch of other robotics companies, and Edsinger ended up at Google as one of the directors of its robotics program, until he left to co-found Hello Robot in 2017.

Meanwhile, since 2007 Kemp has been a robotics professor at Georgia Tech, where he runs the Healthcare Robotics Lab. Kemp’s lab was one of the 11 PR2 beta sites, giving him early experience with a ginormous mobile manipulator. Much of the research that Kemp has spent the last decade on involves robots providing assistance to untrained users, often through direct physical contact, and frequently either in their own homes or in a home environment. We should mention that the Georgia Tech PR2 is still going, most recently doing some clever material classification work in a paper for IROS later this year.

Photo: Hello Robot

Hello Robot co-founder and CEO Aaron Edsinger says that, although Stretch is currently a research platform, he hopes to see the robot deployed in home environments, adding that the “impact we want to have is through robots that are helpful to people in society.”

So with all that in mind, where’d Hello Robot come from? As it turns out, both Edsinger and Kemp were in Rodney Brooks’ group at MIT, so it’s perhaps not surprising that they share some of the same philosophies about what robots should be and what they should be used for. After collaborating on a variety of projects over the years, in 2017 Edsinger was thinking about his next step after Google when Kemp stopped by to show off some video of a new robot prototype that he’d been working on—the prototype for Stretch. “As soon as I saw it, I knew that was exactly the kind of thing I wanted to be working on,” Edsinger told us. “I’d become frustrated with the complexity of the robots being built to do manipulation in home environments and around people, and it solved a lot of problems in an elegant way.”

For Kemp, Stretch is an attempt to get everything he’s been teaching his robots out of his lab at Georgia Tech and into the world where it can actually be helpful to people. “Right from the beginning, we were trying to take our robots out to real homes and interact with real people,” says Kemp. Georgia Tech’s PR2, for example, worked extensively with Henry and Jane Evans, helping Henry (a quadriplegic) regain some of the bodily autonomy he had lost. With the assistance of the PR2, Henry was able to keep himself comfortable for hours without needing a human caregiver to be constantly with him. “I felt like I was making a commitment in some ways to some of the people I was working with,” Kemp told us. “But 10 years later, I was like, where are these things? I found that incredibly frustrating. Stretch is an effort to try to push things forward.”

A robot you can put in the backseat of a car
One way to put Stretch in context is to think of it almost as a reaction to the kitchen sink philosophy of the PR2. Where the PR2 was designed to be all the robot anyone could ever need (plus plenty of robot that nobody really needed) embodied in a piece of hardware that weighs 225 kilograms and cost nearly half a million dollars, Stretch is completely focused on being just the robot that is actually necessary in a form factor that’s both much smaller and affordable. The entire robot weighs a mere 23 kg in a footprint that’s just a 34 cm square. As you can see from the video, it’s small enough (and safe enough) that it can be moved by a child. The cost? At $17,950 apiece—or a bit less if you buy a bunch at once—Stretch costs a fraction of what other mobile manipulators sell for.

It might not seem like size or weight should be that big of an issue, but it very much is, explains Maya Cakmak, a robotics professor at the University of Washington, in Seattle. Cakmak worked with PR2 and Henry Evans when she was at Willow Garage, and currently has access to both a PR2 and a Fetch research robot. “When I think about my long term research vision, I want to deploy service robots in real homes,” Cakmak told us. Unfortunately, it’s the robots themselves that have been preventing her from doing this—both the Fetch and the PR2 are large enough that moving them anywhere requires a truck and a lift, which also limits the home that they can be used in. “For me, I felt immediately that Stretch is very different, and it makes a lot of sense,” she says. “It’s safe and lightweight, you can probably put it in the backseat of a car.” For Cakmak, Stretch’s size is the difference between being able to easily take a robot to the places she wants to do research in, and not. And cost is a factor as well, since a cheaper robot means more access for her students. “I got my refurbished PR2 for $180,000,” Cakmak says. “For that, with Stretch I could have 10!”

“I felt immediately that Stretch is very different. It’s safe and lightweight, you can probably put it in the backseat of a car. I got my refurbished PR2 for $180,000. For that, with Stretch I could have 10!”
—Maya Cakmak, University of Washington

Of course, a portable robot doesn’t do you any good if the robot itself isn’t sophisticated enough to do what you need it to do. Stretch is certainly a compromise in functionality in the interest of small size and low cost, but it’s a compromise that’s been carefully thought out, based on the experience that Edsinger has building robots and the experience that Kemp has operating robots in homes. For example, most mobile manipulators are essentially multi-degrees-of-freedom arms on mobile bases. Stretch instead leverages its wheeled base to move its arm in the horizontal plane, which (most of the time) works just as well as an extra DoF or two on the arm while saving substantially on weight and cost. Similarly, Stretch relies almost entirely on one sensor, an Intel RealSense D435i on a pan-tilt head that gives it a huge range of motion. The RealSense serves as a navigation camera, manipulation camera, a 3D mapping system, and more. It’s not going to be quite as good for a task that might involve fine manipulation, but most of the time it’s totally workable and you’re saving on cost and complexity.

Stretch has been relentlessly optimized to be the absolutely minimum robot to do mobile manipulation in a home or workplace environment. In practice, this meant figuring out exactly what it was absolutely necessary for Stretch to be able to do. With an emphasis on manipulation, that meant defining the workspace of the robot, or what areas it’s able to usefully reach. “That was one thing we really had to push hard on,” says Edsinger. “Reachability.” He explains that reachability and a small mobile base tend not to go together, because robot arms (which tend to weigh a lot) can cause a small base to tip, especially if they’re moving while holding a payload. At the same time, Stretch needed to be able to access both countertops and the floor, while being able to reach out far enough to hand people things without having to be right next to them. To come up with something that could meet all those requirements, Edsinger and Kemp set out to reinvent the robot arm.

Stretch’s key innovation: a stretchable arm
The design they came up with is rather ingenious in its simplicity and how well it works. Edsinger explains that the arm consists of five telescoping links: one fixed and four moving. They are constructed of custom carbon fiber, and are driven by a single motor, which is attached to the robot’s vertical pole. The strong, lightweight structure allows the arm to extend over half a meter and hold up to 1.5 kg. Although the company has a patent pending for the design, Edsinger declined to say whether the links are driven by a belt, cables, or gears. “We don’t want to disclose too much of the secret sauce [with regard to] the drive mechanism.” He added that the arm was “one of the most significant engineering challenges on the robot in terms of getting the desired reach, compactness, precision, smoothness, force sensitivity, and low cost to all happily coexist.”

Photo: Hello Robot

Stretch’s arm consists of five telescoping links constructed of custom carbon fiber, and are driven by a single motor, which is attached to the robot’s vertical pole, minimizing weight and inertia. The arm has a reach of over half a meter and can hold up to 1.5 kg.

Another interesting features of Stretch is its interface with the world—its gripper. There are countless different gripper designs out there, each and every one of which is the best at gripping some particular subset of things. But making a generalized gripper for all of the stuff that you’d find in a home is exceptionally difficult. Ideally, you’d want some sort of massive experimental test program where thousands and thousands of people test out different gripper designs in their homes for long periods of time and then tell you which ones work best. Obviously, that’s impractical for a robotics startup, but Kemp realized that someone else was already running the study for him: Amazon.

“I had this idea that there are these assistive grabbers that people with disabilities use to grasp objects in the real world,” he told us. Kemp went on Amazon’s website and looked at the top 10 grabbers and the reviews from thousands of users. He then bought a bunch of different ones and started testing them. “This one [Stretch’s gripper], I almost didn’t order it, it was such a weird looking thing,” he says. “But it had great reviews on Amazon, and oh my gosh, it just blew away the other grabbers. And I was like, that’s it. It just works.”

Stretch’s teleoperated and autonomous capabilities
As with any robot intended to be useful outside of a structured environment, hardware is only part of the story, and arguably not even the most important part. In order for Stretch to be able to operate out from under the supervision of a skilled roboticist, it has to be either easy to control, or autonomous. Ideally, it’s both, and that’s what Hello Robot is working towards, although things didn’t start out that way, Kemp explains. “From a minimalist standpoint, we began with the notion that this would be a teleoperated robot. But in the end, you just don’t get the real power of the robot that way, because you’re tied to a person doing stuff. As much as we fought it, autonomy really is a big part of the future for this kind of system.”

Here’s a look at some of Stretch’s teleoperated capabilities. We’re told that Stretch is very easy to get going right out of the box, although this teleoperation video from Hello Robot looks like it’s got a skilled and experienced user in the loop:

For such a low-cost platform, the autonomy (even at this early stage) is particularly impressive:

Since it’s not entirely clear from the video exactly what’s autonomous, here’s a brief summary of a couple of the more complex behaviors that Kemp sent us:

Object grasping: Stretch uses its 3D camera to find the nearest flat surface using a virtual overhead view. It then segments significant blobs on top of the surface. It selects the largest blob in this virtual overhead view and fits an ellipse to it. It then generates a grasp plan that makes use of the center of the ellipse and the major and minor axes. Once it has a plan, Stretch orients its gripper, moves to the pre-grasp pose, moves to the grasp pose, closes its gripper based on the estimated object width, lifts up, and retracts.
Mapping, navigating, and reaching to a 3D point: These demonstrations all use FUNMAP (Fast Unified Navigation, Manipulation and Planning). It’s all novel custom Python code. Even a single head scan performed by panning the 3D camera around can result in a very nice 3D representation of Stretch’s surroundings that includes the nearby floor. This is surprisingly unusual for robots, which often have their cameras too low to see many interesting things in a human environment. While mapping, Stretch selects where to scan next in a non-trivial way that considers factors such as the quality of previous observations, expected new observations, and navigation distance. The plan that Stretch uses to reach the target 3D point has been optimized for navigation and manipulation. For example, it finds a final robot pose that provides a large manipulation workspace for Stretch, which must consider nearby obstacles, including obstacles on the ground.
Object handover: This is a simple demonstration of object handovers. Stretch performs Cartesian motions to move its gripper to a body-relative position using a good motion heuristic, which is to extend the arm as the last step. These simple motions work well due to the design of Stretch. It still surprises me how well it moves the object to comfortable places near my body, and how unobtrusive it is. The goal point is specified relative to a 3D frame attached to the person’s mouth estimated using deep learning models (shown in the RViz visualization video). Specifically, Stretch targets handoff at a 3D point that is 20 cm below the estimated position of the mouth and 25 cm away along the direction of reaching.

Much of these autonomous capabilities come directly from Kemp’s lab, and the demo code is available for anyone to use. (Hello Robot says all of Stretch’s software is open source.)

Photo: Hello Robot

Hello Robot co-founder and CEO Aaron Edsinger says Stretch is designed to work with people in homes and workplaces and can be teleoperated to do a variety of tasks, including picking up toys, removing laundry from a dryer, and playing games with kids.

As of right now, Stretch is very much a research platform. You’re going to see it in research labs doing research things, and hopefully in homes and commercial spaces as well, but still under the supervision of professional roboticists. As you may have guessed, though, Hello Robot’s vision is a bit broader than that. “The impact we want to have is through robots that are helpful to people in society,” Edsinger says. “We think primarily in the home context, but it could be in healthcare, or in other places. But we really want to have our robots be impactful, and useful. To us, useful is exciting.” Adds Kemp: “I have a personal bias, but we’d really like this technology to benefit older adults and caregivers. Rather than creating a specialized assistive device, we want to eventually create an inexpensive consumer device for everyone that does lots of things.”

Neither Edsinger nor Kemp would say much more on this for now, and they were very explicit about why—they’re being deliberately cautious about raising expectations, having seen what’s happened to some other robotics companies over the past few years. Without VC funding (Hello Robot is currently bootstrapping itself into existence), Stretch is being sold entirely on its own merits. So far, it seems to be working. Stretch robots are already in a half dozen research labs, and we expect that with today’s announcement, we’ll start seeing them much more frequently.

This article appears in the October 2020 print issue as “A Robot That Keeps It Simple.” Continue reading

Posted in Human Robots

#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