Tag Archives: robotics
#439416 Neuro-evolutionary robotics: A gap ...
Neuro-evolutionary robotics is an attractive approach to realize collective behaviors for swarms of robots. Despite the large number of studies that have been devoted to it and although many methods and ideas have been proposed, empirical evaluations and comparative analyses are rare. Continue reading
Posted in Human Robots
#439406 Dextrous Robotics Wants To Move Boxes ...
Hype aside, there aren’t necessarily all that many areas where robots have the potential to step into an existing workflow and immediately provide a substantial amount of value. But one of the areas that we have seen several robotics companies jump into recently is box manipulation—specifically, using robots to unload boxes from the back of a truck, ideally significantly faster than a human. This is a good task for robots because it plays to their strengths: you can work in a semi-structured and usually predictable environment, speed, power, and precision are all valued highly, and it’s not a job that humans are particularly interested in or designed for.
One of the more novel approaches to this task comes from Dextrous Robotics, a Memphis TN-based startup led by Evan Drumwright. Drumwright was a professor at GWU before spending a few years at the Toyota Research Institute and then co-founding Dextrous in 2019 with an ex-student of his, Sam Zapolsky. The approach that they’ve come up with is to do box manipulation without any sort of suction, or really any sort of grippers at all. Instead, they’re using what can best be described as a pair of moving arms, each gripping a robotic chopstick.
We can pick up basically anything using chopsticks. If you're good with chopsticks, you can pick up individual grains of rice, and you can pick up things that are relatively large compared to the scale of the chopsticks. Your imagination is about the limit, so wouldn't it be cool if you had a robot that could manipulate things with chopsticks? —Evan Drumwright
It definitely is cool, but are there practical reasons why using chopsticks for box manipulation is a good idea? Of course there are! The nice thing about chopsticks is that they really can grip almost anything (even if you scale them up), making them especially valuable in constrained spaces where you’ve got large disparities in shapes and sizes and weights. They’re good for manipulation, too, able to nudge and reposition things with precision. And while Dextrous is initially focused on a trailer unloading task, having this extra manipulation capability will allow them to consider more difficult manipulation tasks in the future, like trailer loading, a task that necessarily happens just as often as unloading does but which is significantly more complicated to robot-ize.
Even though there are some clear advantages to Dextrous’ chopstick technique, there are disadvantages as well, and the biggest one is likely that it’s just a lot harder to use a manipulation technique like this. “The downside of the chopsticks approach is, as any human will tell you, you need some sophisticated control software to be able to operate,” Drumwright tells us. “But that’s part of what we bring to the game: not just a clever hardware design, but the software to operate it, too.”
Meanwhile, what we’ve seen so far from other companies in this space is pretty consistent use of suction systems for box handling. If you have a flat, non-permeable surface (as with most boxes), suction can work quickly and reliably and with a minimum of fancy planning. However, suction has limits form of manipulation, because it’s inherently so sticky, meaning that it can be difficult and/or time consuming to do anything with precision. Other issues with suction include its sensitivity to temperature and moisture, its propensity to ingest all the dirt it possibly can, and the fact that you need to design the suction array based on the biggest and heaviest things that you anticipate having to deal with. That last thing is a particular problem because if you also want to manipulate smaller objects, you’re left trying to do so with a suction array that’s way bigger than you’d like it to be. This is not to say that suction is inferior in all cases, and Drumwright readily admits that suction will probably prove to be a good option for some specific tasks. But chopstick manipulation, if they can get it to work, will be a lot more versatile.
Photo: Dextrous Robotics
Dextrous Robotics co-founders Evan Drumwright and Sam Zapolsky.
I think there's a reason that nature has given us hands. Nature knows how to design suction devices—bats have it, octopi have it, frogs have it—and yet we have hands. Why? Hands are a superior instrument. And so, that's why we've gone down this road. I personally believe, based on billions of years of evolution, that there's a reason that manipulation is superior and that that technology is going to win out. —Evan Drumwright
Part of Dextrous’ secret sauce is an emphasis on simulation. Hardware is hard, so ideally, you want to make one thing that just works the first time, rather than having to iterate over and over. Getting it perfect on the first try is probably unrealistic, but the better you can simulate things in advance, the closer you can get. “What we’ve been able to do is set up our entire planning perception and control system so that it looks exactly like it does when that code runs on the real robot,” says Drumwright. “When we run something on the simulated robot, it agrees with reality about 95 percent of the time, which is frankly unprecedented.” Using very high fidelity hardware modeling, a real time simulator, and software that can directly transfer between sim and real, Dextrous is able to confidently model how their system performs even on notoriously tricky things to simulate, like contact and stiction. The idea is that the end result will be a system that can be developed faster while performing more complex tasks better than other solutions.
We were also wondering why this system uses smooth round chopsticks rather than something a little bit grippier, like chopsticks with a square cross section, and maybe with some higher friction something on the inside surface. Drumwright explains that the advantage of the current design is that it’s symmetrical around its rotational axis, meaning that you only need five degrees of freedom to fully control it. “What that means practically is that things can get a whole lot simpler—the control algorithms get simpler, the inverse kinematics algorithms get simpler, and importantly the number of motors that we need to drive in the robot goes down.”
Screenshot: Dextrous Robotics
Simulated version of Dextrous Robotics’ hardware.
Dextrous took seed funding 18 months ago, and since then they’ve been working on both the software and hardware for their system as well as finding the time to score an NSF SBIR phase 1 grant. The above screenshot shows the simulation of the hardware they’re working towards (chopstick manipulators on two towers that can move laterally), while the Franka Panda arms are what they’re using to validate their software in the meantime. New hardware should be done imminently, and over the next year, Dextrous is looking forward to conducting paid pilots with real customers. Continue reading
Posted in Human Robots
#439386 Zebra Technologies To Acquire Fetch ...
A company called Zebra Technologies announced this morning that it intends to acquire Fetch Robotics for an impressive $305 million.
Fetch is best known for its autonomous mobile robots (AMRs) for warehouses and boasts “the largest portfolio of AMRs in the industry,” and we’re particular fans of its one-armed mobile manipulator for research. Zebra, meanwhile, does stuff with barcodes (get it?), and has been actively investing in robotics companies with a goal of increasing its footprint in the intelligent industrial automation space.
According to the press release, the acquisition “will provide an innovative offering that drives greater efficiencies and higher ROI through better orchestration of technology and people.” We have no idea what that means, but fortunately, we’ve been able to speak with both Fetch and Zebra for details about the deal.
Fetch Robotics’ $305 million purchase price includes $290 million in cash to snap up the 95% of Fetch that Zebra doesn’t already own—Zebra had already invested in Fetch through Zebra Ventures, which also has Locus Robotics and Plus One robotics in its portfolio. There are still some “customary closing conditions” and regulatory approvals that need to happen, so everything isn’t expected to get wrapped up for another month or so. And when it does, it will in some ways mark the end of a robotics story that we’ve been following for the better part of a decade.
Fetch Robotics was founded in early 2015 by the same team of robot experts who had founded Unbounded Robotics just two years before. Melonee Wise, Michael Ferguson, Derek King, and Eric Diehr all worked at Willow Garage, and Unbounded was a mobile manipulation-focused spin-off of Willow that didn’t pan out for reasons that are still not super clear. But in any case, Fetch was a fresh start that allowed Wise, Ferguson, King, and Diehr to fully develop their concept for an intelligent, robust, and efficient autonomous mobile robotic system.
Most of what Fetch Robotics does is warehouse logistics—moving stuff from one place to another so that humans don’t have to. Their autonomous mobile robots work outside of warehouses as well, most recently by providing disinfection services for places like airports. There are plenty of other companies in the larger AMR space, but from what we understand, what Fetch has been doing for the last five years has been consistently state of the art.
This is why Fetch makes sense as an acquisition target, I think: they’ve got exceptional technology in an area (fulfillment, mostly) that has been undergoing a huge amount of growth and where robotics has an enormous opportunity. But what about Zebra Technologies? As far as I can make out, Zebra is one of those companies that you’ve probably never heard of but is actually enormous and everywhere. According to Fortune, as of 2020 they were the 581st biggest company in the world (just behind Levi Strauss) with a market value of $25 billion. While Zebra was founded in 1969, the Zebra-ness didn’t come into play until the early 1980s when they started making barcode printers and scanners. They got into RFID in the early 2000s, and then acquired Motorola’s enterprise unit in 2014, giving Zebra a huge mobile technology portfolio.
To find out where robots fit into all of this, and to learn more about what this means for Fetch, we spoke with Melonee Wise, CEO of Fetch, and Jim Lawton, Vice President and General Manager of Robotics Automation at Zebra.
IEEE Spectrum: Can you tell us about Zebra’s background and interest in robotics?
Jim Lawton: Zebra is a combination of companies that have come together over time. Historically, we were a printing company that made barcode labels, and then we acquired a mobile computing business from Motorola, and today we have a variety of devices that do sensing, analyzing, and acting—we’ve been getting increasingly involved in automation in general.
A lot of our major customers are retailers, warehousing, transportation and logistics, or healthcare, and what we’ve heard a lot lately is that there is an increased pressure towards trying to figure out how to run a supply chain efficiently. Workflows have gotten much more complicated and many of our customers don't feel like they're particularly well equipped to sort through those challenges. They understand that there's an opportunity to do something significant with robots, but what does that look like? What are the right strategies? And they're asking us for help.
There are lots of AMR companies out there doing things that superficially seem similar, but what do you feel is special about Fetch?
Jim Lawton: I was at Universal Robots for a while, and at Rethink Robotics for a number of years, and designing and building robots and bringing them to market is really, really hard. The only way to pull it off is with an amazing team, and Melonee has done an extraordinarily outstanding job, pulling together a world class robotics team.
We had invested in Fetch Robotics a couple of years ago, so we've been working pretty closely together already. We invest in companies in part so that we can educate ourselves, but it's also an opportunity to see whether we’re a good fit with each other. Zebra is a technology and engineering oriented company, and Fetch is as well. With the best team, and the best robots, we just think there’s an outstanding opportunity that we haven’t necessarily found with other AMR companies.
What about for Fetch? Why is Zebra a good fit?
Melonee Wise: Over the last couple of years we have been slowly expanding the devices that we want to connect to, and the software ecosystems that we want to connect to, and Zebra has provided a lot of that synergy. We're constantly asked, can we get a robot to do something if we scan a barcode, or can we press a button on a tablet, and have a robot appear, things like that. Being able to deliver these kinds of end to end, fully encapsulated solutions that go beyond the robots and really solve the problems that customers are looking to solve—Zebra helps us do that.
And there's also an opportunity for us as a robotics startup to partner with a larger company to help us scale much more rapidly. That’s the other thing that’s really exciting for us—Zebra has a very strong business in warehousing and logistics. They’re an industry leader, and I think they can really help us get to the next level as a company.
Does that represent a transition for AMRs from just moving things from one place to another to integrating with all kinds of other warehouse systems?
Melonee Wise: For a decade or more, people have been talking about Industry 4.0 and how it's going to change the world and revolutionize manufacturing, but as a community we’ve struggled to execute on that goal for lots of reasons. We've had what people might call islands of automation: siloed pieces of automation that are doing their thing by themselves. But if they have to talk to each other, that's a bridge too far.
But in many ways automation technology is now getting mature enough through the things that we’ve seen in software for a long time, like APIs, interconnected services, and cloud platforms. Zebra has been working on that independently for a long time as part of their business, and so bringing our two businesses together to build these bridges between islands of automation is why it made sense for us to come together at this point in time.
If you go back far enough, Fetch has its origins in Willow Garage and ROS, and I know that Fetch still makes substantial software contributions back to the ROS community. Is that something you’ll be able to continue?
Melonee Wise: Our participation in the open source community is still very important, and I think it’s going to continue to be important. A lot of robotics is really about getting great talent, and open source is one way that we connect to that talent and participate in the larger ecosystem and draw value from it. There are also lots of great tools out there in the open source community that Fetch uses and contributes to. And I think those types of projects that are not core to our IP but give us value will definitely be things that we continue to participate in.
What will happen to the Fetch mobile manipulator that I know a lot of labs are currently using for research?
Melonee Wise: We're committed to continuing to support our existing customers and I think that there’s still a place for the research product going forward.
What do you think are the biggest challenges for AMRs right now?
Melonee Wise: One thing that I think is happening in the industry is that the safety standards are now coming into play. In December of last year the first official autonomous mobile robot safety standards were released, and not everyone was ready for that, but Fetch has been at the front of this for a long time. It took about four years to develop the AMR safety standard, and getting to an understanding of what safe actually means and how you implement those safety measures. It’s common for safety standards to lag behind technology, but customers have been asking more and more, “well how do I know that your robots are safe?” And so I think what we're going to see is that these safety standards are going to have differing effects on different companies, based on how thoughtful they've been about safety through the design and implementation of their technology,
What have you learned, or what has surprised you about your industry now that we’re a year and a half into the pandemic?
Melonee Wise: One of the more interesting things to me was that it was amazing how quickly the resistance to the cloud goes away when you have to deploy things remotely during a pandemic. Originally customers weren't that excited about the cloud and wanted to do everything on site, but once the pandemic hit they switched their point of view on the technology pretty quickly, which was nice to see.
Jim Lawton: The amount of interest that we've seen in robots and automation in general has skyrocketed over the last year. In particular we’re hearing from companies that are not well equipped to deal with their automation needs, and the pandemic has just made it so much more clear to them that they have to do something. I think we're going to see a renaissance within some of these spaces because of their investment in robotic technologies. Continue reading
Posted in Human Robots
#439378 SoftBank Stops Making Pepper Robots, ...
Reuters is reporting that SoftBank stopped manufacturing Pepper robots at some point last year due to low demand, and by September, will cut about half of the 330 positions at SoftBank Robotics Europe in France. Most of the positions will be in Q&A, sales, and service, which hopefully leaves SoftBank Robotics’ research and development group mostly intact. But the cuts reflect poor long-term sales, with SoftBank Robotics Europe having lost over 100 million Euros in the past three years, according to French business news site JDN. Speaking with Nikkei, SoftBank said that this doesn’t actually mean a permanent end for Pepper, and that they “plan to resume production if demand recovers.” But things aren’t looking good.
Reuters says that “only” 27,000 Peppers were produced, but that sure seems like a lot of Peppers to me. Perhaps too many—a huge number of Peppers were used by SoftBank itself in its retail stores, and a hundred at once were turned into a cheerleading squad for the SoftBank Hawks baseball team because of the pandemic. There’s nothing wrong with either of those things, but it’s hard to use them to gauge how successful Pepper has actually been.
I won’t try to argue that Pepper would necessarily have been commercially viable in the long(er) term, since it’s a very capable robot in some ways, but not very capable in others. For example, Pepper has arms and hands with individually articulated fingers, but the robot can’t actually do much in the way of useful grasping or manipulation. SoftBank positioned Pepper as a robot that can attract attention and provide useful, socially interactive information in public places. Besides SoftBank’s own stores, Peppers have been used in banks, malls, airports, and other places of that nature. A lot of what Pepper seems to have uniquely offered was novelty, though, which ultimately may not be sustainable for a commercial robot, because at some point, the novelty just wears off and you’re basically left with a very cool looking (but expensive) kiosk.
Having said all that, the sheer number of Peppers that SoftBank put out in the world could be one of the most significant impacts that the robot has had. The fact that Pepper was able to successfully operate for long enough, and in enough places, that it even had a chance to stop becoming novel and instead become normal is an enormous achievement for Pepper specifically as well as for social robots more broadly. Angelica Lim, who worked with Pepper at SoftBank Robotics Europe for three years before founding the Rosie Lab at SFU, shared some perspective with us on this:
There has never been a robot with the ambition of Pepper. Its mission was huge—be adaptable and robust to different purposes and locations: loud sushi shops, quiet banks, and hospitals that change from hour to hour. Compare that with Alexa which has a pretty stable and quiet environment—the home. On top of that, the robot needed to respond to different ages, cultures, countries and languages. The only thing I can think of that comes close is the smartphone, and the expectation for it is much lower compared to the humanoid Pepper. Ten years ago, it was unthinkable that we could leave a robot on “in the wild” for days, weeks, months and years, and yet Pepper did it thanks to the team at SoftBank Robotics.
Peppers are still being used in education today, from elementary schools and high schools to research labs in North America, Asia and Europe. The next generation will grow up programming these, like they did with the Apple personal computer. I’m confident it’s just the next step to technology that adapts to us as humans rather than the other way around.
Pepper has been an amazing platform for HRI research as well as for STEM education more broadly, and our hope is that Pepper will continue to be impactful in those ways, whether or not any more of these robots are ever made. We also hope that SoftBank does whatever is necessary to make sure that Peppers remain useful and accessible well into the future in both software and hardware. But perhaps we’re being too pessimistic here—this is certainly not good news, but despite how it looks we don’t know for sure that it’s catastrophic for Pepper. All we can do is wait and see what happens at SoftBank Robotics Europe over the next six months, and hope that Pepper continues to get the support that it deserves. Continue reading
Posted in Human Robots
#439292 New processes for automated fabrication ...
Researchers from the Singapore University of Technology and Design (SUTD) have developed novel techniques, known as Automated Fiber Embedding (AFE), to produce complex fiber and silicone composite structures for soft robotics applications. Their work was published in IEEE Robotics and Automation Letters. Continue reading
Posted in Human Robots