Tag Archives: with
#439509 What’s Going on With Amazon’s ...
Amazon’s innovation blog recently published a post entitled “New technologies to improve Amazon employee safety,” which highlighted four different robotic systems that Amazon’s Robotics and Advanced Technology teams have been working on. Three of these robotic systems are mobile robots, which have been making huge contributions to the warehouse space over the past decade. Amazon in particular was one of the first (if not the first) e-commerce companies to really understand the fundamental power of robots in warehouses, with their $775 million acquisition of Kiva Systems’ pod-transporting robots back in 2012.
Since then, a bunch of other robotics companies have started commercially deploying robots in warehouses, and over the past five years or so, we’ve seen some of those robots develop enough autonomy and intelligence to be able to operate outside of restricted, highly structured environments and work directly with humans. Autonomous mobile robots for warehouses is now a highly competitive sector, with companies like Fetch Robotics, Locus Robotics, and OTTO Motors all offering systems that can zip payloads around busy warehouse floors safely and efficiently.
But if we’re to take the capabilities of the robots that Amazon showcased over the weekend at face value, the company appears to be substantially behind the curve on warehouse robots.
Let’s take a look at the three mobile robots that Amazon describes in their blog post:
“Bert” is one of Amazon’s first Autonomous Mobile Robots, or AMRs. Historically, it’s been difficult to incorporate robotics into areas of our facilities where people and robots are working in the same physical space. AMRs like Bert, which is being tested to autonomously navigate through our facilities with Amazon-developed advanced safety, perception, and navigation technology, could change that. With Bert, robots no longer need to be confined to restricted areas. This means that in the future, an employee could summon Bert to carry items across a facility. In addition, Bert might at some point be able to move larger, heavier items or carts that are used to transport multiple packages through our facilities. By taking those movements on, Bert could help lessen strain on employees.
This all sounds fairly impressive, but only if you’ve been checked out of the AMR space for the last few years. Amazon is presenting Bert as part of the “new technologies” they’re developing, and while that may be the case, as far as we can make out these are very much technologies that seem to be new mostly just to Amazon and not really to anyone else. There are any number of other companies who are selling mobile robot tech that looks to be significantly beyond what we’re seeing here—tech that (unless we’re missing something) has already largely solved many of the same technical problems that Amazon is working on.
We spoke with mobile robot experts from three different robotics companies, none of whom were comfortable going on record (for obvious reasons), but they all agreed that what Amazon is demonstrating in these videos appears to be 2+ years behind the state of the art in commercial mobile robots.
We’re obviously seeing a work in progress with Bert, but I’d be less confused if we were looking at a deployed system, because at least then you could make the argument that Amazon has managed to get something operational at (some) scale, which is much more difficult than a demo or pilot project. But the slow speed, the careful turns, the human chaperones—other AMR companies are way past this stage.
Kermit is an AGC (Autonomously Guided Cart) that is focused on moving empty totes from one location to another within our facilities so we can get empty totes back to the starting line. Kermit follows strategically placed magnetic tape to guide its navigation and uses tags placed along the way to determine if it should speed up, slow down, or modify its course in some way. Kermit is further along in development, currently being tested in several sites across the U.S., and will be introduced in at least a dozen more sites across North America this year.
Most folks in the mobile robots industry would hesitate to call Kermit an autonomous robot at all, which is likely why Amazon doesn’t refer to it as such, instead calling it a “guided cart.” As far as I know, pretty much every other mobile robotics company has done away with stuff like magnetic tape in favor of map-based natural-feature localization (a technology that has been commercially available for years), because then your robots can go anywhere in a mapped warehouse, not just on these predefined paths. Even if you have a space and workflow that never ever changes, busy warehouses have paths that get blocked for one reason or another all the time, and modern AMRs are flexible enough to plan around those paths to complete their tasks. With these autonomous carts that are locked to their tapes, they can’t even move over a couple of feet to get around an obstacle.
I have no idea why this monstrous system called Scooter is the best solution for moving carts around a warehouse. It just seems needlessly huge and complicated, especially since we know Amazon already understands that a great way of moving carts around is by using much smaller robots that can zip underneath a cart, lift it up, and carry it around with them. Obviously, the Kiva drive units only operate in highly structured environments, but other AMR companies are making this concept work on the warehouse floor just fine.
Why is Amazon at “possibilities” when other companies are at commercial deployments?
I honestly just don’t understand what’s happening here. Amazon has (I assume) a huge R&D budget at its disposal. It was investing in robotic technology for e-commerce warehouses super early, and at an unmatched scale. Even beyond Kiva, Amazon obviously understood the importance of AMRs several years ago, with its $100+ million acquisition of Canvas Technology in 2019. But looking back at Canvas’ old videos, it seems like Canvas was doing in 2017 more or less what we’re seeing Amazon’s Bert robot doing now, nearly half a decade later.
We reached out to Amazon Robotics for comment and sent them a series of questions about the robots in these videos. They sent us this response:
The health and safety of our employees is our number one priority—and has been since day one. We’re excited about the possibilities robotics and other technology can play in helping to improve employee safety.
Hmm.
I mean, sure, I’m excited about the same thing, but I’m still stuck on why Amazon is at possibilities, while other companies are at commercial deployments. It’s certainly possible that the sheer Amazon-ness of Amazon is a significant factor here, in the sense that a commercial deployment for Amazon is orders of magnitude larger and more complex than any of the AMR companies that we’re comparing them to are dealing with. And if Amazon can figure out how to make (say) an AMR without using lidar, it would make a much more significant difference for an in-house large-scale deployment relative to companies offering AMRs as a service.
For another take on what might be going on with this announcement from Amazon, we spoke with Matt Beane, who got his PhD at MIT and studies robotics at UCSB’s Technology Management Program. At the ACM/IEEE International Conference on Human-Robot Interaction (HRI) last year, Beane published a paper on the value of robots as social signals—that is, organizations get valuable outcomes from just announcing they have robots, because this encourages key audiences to see the organization in favorable ways. “My research strongly suggests that Amazon is reaping signaling value from this announcement,” Beane told us. There’s nothing inherently wrong with signaling, because robots can create instrumental value, and that value needs to be communicated to the people who will, ideally, benefit from it. But you have to be careful: “My paper also suggests this can be a risky move,” explains Beane. “Blowback can be pretty nasty if the systems aren’t in full-tilt, high-value use. In other words, it works only if the signal pretty closely matches the internal reality.”
There’s no way for us to know what the internal reality at Amazon is. All we have to go on is this blog post, which isn’t much, and we should reiterate that there may be a significant gap between what the post is showing us about Amazon’s mobile robots and what’s actually going on at Amazon Robotics. My hope is what we’re seeing here is primarily a sign that Amazon Robotics is starting to scale things up, and that we’re about to see them get a lot more serious about developing robots that will help make their warehouses less tedious, safer, and more productive. Continue reading
Posted in Human Robots
#439465 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.
Dextrous Robotics co-founders Evan Drumwright and Sam Zapolsky.
Photo: Dextrous Robotics
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.”
Simulated version of Dextrous Robotics’ hardware.
Screenshot: Dextrous Robotics
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
#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
#439374 A model to predict how much humans and ...
Researchers at University of Michigan have recently developed a bi-directional model that can predict how much both humans and robotic agents can be trusted in situations that involve human-robot collaboration. This model, presented in a paper published in IEEE Robotics and Automation Letters, could help to allocate tasks to different agents more reliably and efficiently. Continue reading
Posted in Human Robots
#439353 What’s Going on With Amazon’s ...
Amazon’s innovation blog recently published a post entitled “New technologies to improve Amazon employee safety,” which highlighted four different robotic systems that Amazon’s Robotics and Advanced Technology teams have been working on. Three of these robotic systems are mobile robots, which have been making huge contributions to the warehouse space over the past decade. Amazon in particular was one of the first (if not the first) e-commerce companies to really understand the fundamental power of robots in warehouses, with their $775 million acquisition of Kiva Systems’ pod-transporting robots back in 2012.
Since then, a bunch of other robotics companies have started commercially deploying robots in warehouses, and over the past five years or so, we’ve seen some of those robots develop enough autonomy and intelligence to be able to operate outside of restricted, highly structured environments and work directly with humans. Autonomous mobile robots for warehouses is now a highly competitive sector, with companies like Fetch Robotics, Locus Robotics, and OTTO Motors all offering systems that can zip payloads around busy warehouse floors safely and efficiently.
But if we’re to take the capabilities of the robots that Amazon showcased over the weekend at face value, the company appears to be substantially behind the curve on warehouse robots.
Let’s take a look at the three mobile robots that Amazon describes in their blog post:
“Bert” is one of Amazon’s first Autonomous Mobile Robots, or AMRs. Historically, it’s been difficult to incorporate robotics into areas of our facilities where people and robots are working in the same physical space. AMRs like Bert, which is being tested to autonomously navigate through our facilities with Amazon-developed advanced safety, perception, and navigation technology, could change that. With Bert, robots no longer need to be confined to restricted areas. This means that in the future, an employee could summon Bert to carry items across a facility. In addition, Bert might at some point be able to move larger, heavier items or carts that are used to transport multiple packages through our facilities. By taking those movements on, Bert could help lessen strain on employees.
This all sounds fairly impressive, but only if you’ve been checked out of the AMR space for the last few years. Amazon is presenting Bert as part of the “new technologies” they’re developing, and while that may be the case, as far as we can make out these are very much technologies that seem to be new mostly just to Amazon and not really to anyone else. There are any number of other companies who are selling mobile robot tech that looks to be significantly beyond what we’re seeing here—tech that (unless we’re missing something) has already largely solved many of the same technical problems that Amazon is working on.
We spoke with mobile robot experts from three different robotics companies, none of whom were comfortable going on record (for obvious reasons), but they all agreed that what Amazon is demonstrating in these videos appears to be 2+ years behind the state of the art in commercial mobile robots.
We’re obviously seeing a work in progress with Bert, but I’d be less confused if we were looking at a deployed system, because at least then you could make the argument that Amazon has managed to get something operational at (some) scale, which is much more difficult than a demo or pilot project. But the slow speed, the careful turns, the human chaperones—other AMR companies are way past this stage.
Kermit is an AGC (Autonomously Guided Cart) that is focused on moving empty totes from one location to another within our facilities so we can get empty totes back to the starting line. Kermit follows strategically placed magnetic tape to guide its navigation and uses tags placed along the way to determine if it should speed up, slow down, or modify its course in some way. Kermit is further along in development, currently being tested in several sites across the U.S., and will be introduced in at least a dozen more sites across North America this year.
Most folks in the mobile robots industry would hesitate to call Kermit an autonomous robot at all, which is likely why Amazon doesn’t refer to it as such, instead calling it a “guided cart.” As far as I know, pretty much every other mobile robotics company has done away with stuff like magnetic tape in favor of map-based natural-feature localization (a technology that has been commercially available for years), because then your robots can go anywhere in a mapped warehouse, not just on these predefined paths. Even if you have a space and workflow that never ever changes, busy warehouses have paths that get blocked for one reason or another all the time, and modern AMRs are flexible enough to plan around those paths to complete their tasks. With these autonomous carts that are locked to their tapes, they can’t even move over a couple of feet to get around an obstacle.
I have no idea why this monstrous system called Scooter is the best solution for moving carts around a warehouse. It just seems needlessly huge and complicated, especially since we know Amazon already understands that a great way of moving carts around is by using much smaller robots that can zip underneath a cart, lift it up, and carry it around with them. Obviously, the Kiva drive units only operate in highly structured environments, but other AMR companies are making this concept work on the warehouse floor just fine.
Why is Amazon at “possibilities” when other companies are at commercial deployments?
I honestly just don’t understand what’s happening here. Amazon has (I assume) a huge R&D budget at its disposal. It was investing in robotic technology for e-commerce warehouses super early, and at an unmatched scale. Even beyond Kiva, Amazon obviously understood the importance of AMRs several years ago, with its $100+ million acquisition of Canvas Technology in 2019. But looking back at Canvas’ old videos, it seems like Canvas was doing in 2017 more or less what we’re seeing Amazon’s Bert robot doing now, nearly half a decade later.
We reached out to Amazon Robotics for comment and sent them a series of questions about the robots in these videos. They sent us this response:
The health and safety of our employees is our number one priority—and has been since day one. We’re excited about the possibilities robotics and other technology can play in helping to improve employee safety.
Hmm.
I mean, sure, I’m excited about the same thing, but I’m still stuck on why Amazon is at possibilities, while other companies are at commercial deployments. It’s certainly possible that the sheer Amazon-ness of Amazon is a significant factor here, in the sense that a commercial deployment for Amazon is orders of magnitude larger and more complex than any of the AMR companies that we’re comparing them to are dealing with. And if Amazon can figure out how to make (say) an AMR without using lidar, it would make a much more significant difference for an in-house large-scale deployment relative to companies offering AMRs as a service.
For another take on what might be going on with this announcement from Amazon, we spoke with Matt Beane, who got his PhD at MIT and studies robotics at UCSB’s Technology Management Program. At the ACM/IEEE International Conference on Human-Robot Interaction (HRI) last year, Beane published a paper on the value of robots as social signals—that is, organizations get valuable outcomes from just announcing they have robots, because this encourages key audiences to see the organization in favorable ways. “My research strongly suggests that Amazon is reaping signaling value from this announcement,” Beane told us. There’s nothing inherently wrong with signaling, because robots can create instrumental value, and that value needs to be communicated to the people who will, ideally, benefit from it. But you have to be careful: “My paper also suggests this can be a risky move,” explains Beane. “Blowback can be pretty nasty if the systems aren’t in full-tilt, high-value use. In other words, it works only if the signal pretty closely matches the internal reality.”
There’s no way for us to know what the internal reality at Amazon is. All we have to go on is this blog post, which isn’t much, and we should reiterate that there may be a significant gap between what the post is showing us about Amazon’s mobile robots and what’s actually going on at Amazon Robotics. My hope is what we’re seeing here is primarily a sign that Amazon Robotics is starting to scale things up, and that we’re about to see them get a lot more serious about developing robots that will help make their warehouses less tedious, safer, and more productive. Continue reading
Posted in Human Robots