Your Blueprint for Switching to Automated Unloading

Welcome, and thanks for joining us today. I'm Shirley Goh, and this is my colleague Grant Aylward. Today, we're talking about the top considerations for when you're thinking about switching to automated unloading. But first, a little bit about ourselves. I work in product marketing at Boston Dynamics with a focus on warehouse automation. Prior to that, I worked at MIT and Boston College in their research spaces. And before that, I was a staff writer and editor at The Boston Globe. Grant, can you tell us a little about yourself. Yeah. Thanks, Shirley. I'm excited for this conversation. My name is Grant Aylward. I'm a product manager on the Stretch team, driving, trailer unloading. I've been in robots for about 10 years now. I was at Rethink Robotics before, and I've been at Boston Dynamics for five years now. When I joined, I thought I was coming to work on Handle. That's what I was expecting. I was telling everyone in my life about that. And I got there on the first day and they said, actually, hey, no, you're working on this other robot called Stretch. It's an evolution from Handle. And of course, you see it and it's like, oh, four wheels do make a bit more sense in this case. But that was an exciting first day and it's been a fun journey since then. Let's jump into automated unloading. So the doctor is essentially like the front door of the warehouse where everything first comes in. But this has been an area that's been neglected when it comes to integrating new technology. Why do you think that is, and why do you think everyone is interested in automating the inbound area now? Yeah, that's a great question. You're right, it is the front door. That material has to get into the warehouse. So one of the big drivers for automating today is that the labor to do that historically has been easier to find than it is today. Today it's quite hard to find that labor and to retain that labor. So that's the pull from the market. The labor part of the story. On the other side, why it's been neglected, is that previously the technology hasn't been there to be able to automate this task. It is a hard task, but today, we do have that technology. It has arrived. So with that push and the pull from the technology and the labor, it's the right time for automating the dock and lots of companies are turning to that. Yeah, we've made so many gains in robotics I feel like in just the last few years alone. Yeah, totally. So that's a good point. I think in two ways-- and Boston Dynamics is really driving some of this forward is that, look our hardware prices are coming down. This includes the price of actuators, price of camera systems. All these things are important to be able to make that hardware that is at a price point that makes sense for the market. So that's being pushed forward. On the other side, the software is getting better. We all know of these evolutions in AI. So all this is making these unstructured tasks that we see in the market, especially with trailer unloading possible with today's technology. So you mentioned labor, and I know that labor has been a challenge for everyone in the last few years, starting with the pandemic. How are companies using technology to address these challenges? Yeah, the question of labor is one of these big driving factors here. The whole warehouse space is seeing a pain point of labor, of attracting labor, retaining it. But the inbound dock is seeing particularly sensitivity there. Unloading is hard. It's a really tough job. And frankly, it's dangerous. We see these big markets, these port cities, Rotterdam, Savannah, even these strategic supply chain cities like Allentown. Already their labor market is really strained to get people into the warehouses. So then people are OK, they get into the warehouses. But then they are told, OK, hey, you're working at the inbound dock. It's really common to see over a month, 100% turnover at the inbound dock. That's retention you really can't plan on for these warehouses. So that is a real pain point, and it's an operational challenge that they are seeking to resolve with automation. Can you speak to some of the costs of that turnover of getting new people, of getting them onboarded again. Yeah, it's a real challenge. Because oftentimes some of these warehouses have very particular ways they go about doing their operational flow. So you have to train these folks up for doing exactly what's needed to be done when you're unloading a trailer. What procedures need to happen, what rates need to be hit, things like that. These all need to be introduced each time someone new comes along so yes, the job of training someone to unload a truck is pretty straightforward, but it can come with lots of different other pieces that make the whole operation work, that those sorts of training pieces are really hard when your turnover is 100%. Now you and I both talk to our customers a good amount. What are some of the anecdotes you've heard about their challenges of labor, and how they've used robots to fill in the gaps. I was down in Savannah recently. When it was really hot days over the summertime, people would come in and they'd be told, OK, go inside this trailer, and it might be 50 pounds boxes and the person would unload maybe 40 or 50 of them and say, you know what, I'm just going to go home today. And they don't come back. And they don't come back. So I think that that is just stuff that you can't plan around for these operations that need to hit their numbers each day. You mentioned, I remember a story with a customer about snowfall in an area where they don't really get it. They're not really prepared for it with the roads, and people just weren't coming in to work, and there were all these boxes to be moved. So they put our robots on the task. They were happy to have them there to address it when people couldn't get to work safely. All right. Sold. So say you're convinced you decided you want to make this transition from manual to automated unloading. How do you assess whether our robot Stretch or some other solution is the right fit for your DC? What are the considerations that you go through? Yeah, there's a few big ones that you should start with. You need to understand safety, right. You need to understand the safety of the vendor that you're looking into, how they're approaching safety. Because that's where everything needs to start, because these robots will be around people. Reliability. You're going to start relying on these new pieces of technology. You have to know that they can reliably be there for you when your operations needs them. Do they work in your freight? There's lots of different freight sizes out there. Freight types. You need to understand if that vendor works well in your operational envelope. And lastly, does it work in your operations and flows. So it might be able to pick the boxes, but is it able to operate with the equipment you have today? Is it able to ensure that boxes are flowing correctly into the downstream automation you have already. Those are the key things. Safety, reliability, does it work with your freight, and does it work with your operational flows. Can you tell us more about that? Yeah. All these warehouses have equipment already. So it's important to understand, does your new piece of automation work with the equipment you have now. So not only the equipment, but the flow of goods once it comes out. I think all those make it important for you to look into. With Stretch, we seamlessly connect to all the major conveyors out there, the Caljan, Stewart Glapat, FMH, all these conveyors we seamlessly connect to. But Stretch also has lots of different placement options. One of the things you learn when you go to a warehouse is that oftentimes folks are told, hey, place boxes in a certain way, long ways down. Or if you have two make sure they're singulated. These sorts of placement options is a feature of Stretch. So folks can go in there and configure, OK, place boxes this way, or ensure that boxes are singulated in this way. So those sorts of ease of configuration, ease of just fitting into your operation, is a big place where stretch stands out against the competition. When I look at videos online and when I walk trade shows, I see so many unloading solutions out there now. There's a lot of possibilities. They all look good in their demos, they look flawless in their videos. What is setting Stretch apart from the rest of these options? What puts it ahead of the pack? Yeah. Their demos and their videos do look great. Stretch is no longer in startup mode. It's no longer in pilot mode. Stretch is a real industrial product out there working for brands you know, brands you trust. They're relying on Stretch. They're trusting in Stretch and Boston Dynamics to ensure their operations keep running smoothly on schedule. So that's what sets us apart. We've moved 25 million cases across the world today. And we also are moving roughly half a million cases every week. These are large numbers. And so it demonstrates our reliability both as the robot but also as the organization. We're able to support these robots across the world. So that way, these brands can trust and rely on us. And you mentioned not being in startup mode. I feel like we also have 30 years of robot research that's come into Stretch. And not everyone, I think, realizes the other robots that have become part of Stretch. That we started actually with Atlas. That a number of years ago, there was a video that was shown of Atlas moving a box, and people wondered, how can I get that into my operations. Can I have a humanoid robot in my operations. And our team went and designed a robot named Handle, which you know well. Robot on two legs and two wheels. And I've seen the videos. I think it was a very fun robot. But our engineering team had another think and decided that wasn't the ideal form factor. And they arrived at the Stretch that you see today. So we've put all of these learnings over the three decades into our robots, and that's the software and the hardware. There are parts of Atlas in there. There are parts of Spot in there. And we're also fortunate to have the backing of Hyundai. Can you speak to that? Yeah, their support has been awesome. When it comes to Stretch building that reliable product, their expertise in how to build the robot and design a robot that can stand the test of time, that expertise has been incredible. It has pushed us forward, and we're looking forward to building upon that expertise into the future, too. And Grant, you mentioned some of the brands that move, Stretch. What are some of these brands that have trusted Stretch to move their massive amounts of freight? Yeah, good question, Shirley. We have a few multi-year. We've been working with some customers for three plus years now, scaled relationships across the world. So these are relationships that we've built that we have deployments for one company all across the US and in the EU. So that's a global partnership that we've built upon. So we have those great partnerships. And that includes people like DHL, people like Autogroup. And we also have great brands of clothes you and I like to wear, things like GAP, Carhartt. So one of the things I like to think about is that when I go into work and I see people walking around, there's a fair amount of material that people are wearing that was moved by Stretch. GAP, it has Old Navy as well, when we go to Europe Autogroup is moving Bonpreu, right. So these are brands that people are wearing every single day that Stretch helped get to the stores, get into their hands, which I really like. And so other brands include Puma. We also have Lidl. They've been a great partner with us already, and we're excited to grow with them. They're moving a lot of different material, and they've been a great operational partner where we get to learn from them, build a product that suits their needs better. It's been an awesome partnership and I'm looking forward to building further on it. Now I think everyone who's joined us today is curious about what deployment looks like. What can you tell us about the experience, from the moment the crates are delivered on site to the moment that the robot is moving boxes. Yeah. The deployment with Stretch happens in days, not months, like we see with traditional automation. So what happens is the crate arrives with Stretch, another crate with lots of the deployment material. Our field application engineers have already done some pre-work to better understand what integration is going to look like. So that material arrives on site on day one. For the next three to four days what folks are doing is integrating that off robot systems, we call it, with the warehouses equipment. So that means with the conveyor, we put our safety arch over the top. We also plug into the telescopic conveyor, so that way Stretch can wirelessly communicate with the conveyor. Move the conveyor forward. Move it back. Stretch can do that wirelessly. There's also a connection to the internet established. So all those configuration things happen in about three to four days. So at that point, Stretch is ready to go and training begins. So training means that we get the folks in who will be both running the robot, but also the folks who will be training other folks as they come in, as new folks get hired to run the robot and also train each other. One of the great things we've heard is that once training is done in a few weeks afterwards, they're able to train new folks to run Stretch in less than an hour. So it's very quick. So anyway, back to deployment. They'll spend three to four days getting everything ready. Then they'll start training. They'll do training for three to four days, depending on how many people are actually seeking to get trained on that deployment. And then we'll finish up with site acceptance test. Well Stretch will run meet the criteria and then folks will leave and the operations are ready to run with the robot at that point. So already by day five Stretch is producing real value to the site. It's unloading boxes. Those boxes are going into the warehouse. Those boxes are being part of their regular business operations. And this is minimal infrastructure that's installed. Is that correct? Yeah. Good point. Good point. Yeah, well Stretch is completely mobile. It has its own air on board. So we're not seeking to add airlines. It has a battery on board so we don't need power lines. What we do when it comes to safety is that we create a LiDAR field that provides that safety envelope for Stretch, but without adding a bunch of infrastructure. So still, the inbound dock is not cluttered with a bunch of new material. So 25 million boxes moved since we began is a lot. But what types of things is Stretch moving? Great question. We're moving lots of different shoes, lots of different shirts. Of course, we're moving a lot of apparel. But we're also moving small appliances. We're moving consumer packaged goods, we're moving consumer electronics and all sorts of retail products. So whenever you go into your favorite big box store, you can see all these different types of material, Stretch is moving all that. So Stretch is able to just get in front of new goods, you press start, and it picks away at all these different sorts of whether they're heavy goods like microwaves or they're small goods like sandals. Now, you mentioned training. When you get on site, are the associates intimidated by the technology, the big hulking robot, or are they eager for new tech to help them in their work? Yeah, right now in the warehouse, there's lots of different robots in there. So I would say that the surprise that folks have when seeing robots in the warehouses has certainly decreased over time. So they're not super surprised about robots. I think oftentimes we worry about people being nervous, oh, this robot's going to take my job. But when we say, hey, this robot's coming in to unload these trailers and you'll be able to manage these robots rather than unload these cases, everyone gets super excited. Because again, this is a job that people don't want to do. It's strenuous. It hurts in the winter time. It's super cold. It's 0 degrees inside this container and maybe the cardboard is falling apart. Or in the summertime, maybe it's 110 degrees and you're moving 50 pounds cases. This is a job people don't want to do. So when they hear that, hey, Stretch is going to come do this job and you get to manage that robot, they're really excited about it. And I'm assuming they don't need to be robot engineers to operate these robots. What is the training like for them? Do they take to it quickly? Is there a learning curve? Yeah, they take to it super quickly. Training is all on the job training. So we do a little bit of classroom training. Say, hey, here are the different materials that you'll recognize around Stretch. Here's the robot. The way to drive the robot into place is a video game console, so people pick that up pretty easily. So it's a few hours of classroom time to say, hey, here are the different pieces. Here's how you should think about it. Here's how you should communicate it later on when you need to train your colleagues. But then it just shifted right to on the job training. And they take to it super easy, Shirley. I think we've spent a lot of time making sure that when folks, come up to use the robot, they don't really know that they're using cutting edge technology. Stretch is on the forefront of robotics, but to them, it feels like a washing machine. You put it in front of the boxes, and you press a big play button, and the robot takes it from there. So it doesn't need information, it doesn't need input added before it gets going on. What sorts of boxes are inside, Stretch takes care of that. So the training is really straightforward. One of my favorite stories I've heard from a customer about training is how they went about it. There is this one site that set up an obstacle course for people to learn to drive the robot. So they set up traffic cones and ramps and other things from around the warehouse to teach people to drive it around. And the robot's got full mobility. It can go in any direction. It can go on ramps. So I thought that was a really fun story. That even though the robot's autonomous, once you really set it to go, it works on its own. People do have to drive it in place, but I think that's a fun thing to do, to drive a robot with a video game controller. Totally. I think one of the big surprises that I had, lots of our customers have is, so you have this big robot, you know it weighs a lot. It's pretty large. But when you use the controller, it's really responsive. OK, I'm going to touch left. Touch right. It responds, right. Because we have that Omni drive base. So we can move in all sorts of directions and we can move there with really fine precision. How about safety? Because even though all these automation solutions are super efficient, they're powerful, they're big, and they're moving. How do you keep people safe in environments like that? Great point. The job of unloading a container means if it's going to work in downstream operations, you have to move quickly. The robot has to unload very quickly. It's not a job that you can take slowly. So you're right. That this large robot is moving fast. So it's not a safe environment for people to be right next to it. At Boston Dynamics, we take safety super seriously. We want to make a big impact on the world. And we know for us to do that, we have to be safe around people. We have to make sure that what the solutions we design are safe for people to be around. That's how we're going to make a big impact on the world. So when it comes to Stretch and trailer unloading, Stretch is able to go into the container or right outside the container on the dock plate. And we have a LiDAR field on the outside that ensures a separation from people to the robot. So we need to make sure that that safety field and safety altogether works day in, day out and is fail safe. That's really important for us. Because look, if we're going to be a reliable piece of automation, this piece of automation is going to run at 2:00 AM. And so people who are tired, people who are whatever, are not going to accidentally be hit by a Stretch, right. So we really design that fail safe when it comes to our safety. So with Stretch, it's like we talked about, that low infrastructure LiDAR field outside the dock door to ensure that safe distance between people and Stretch when it's operating. And that's not physical fencing. It's virtual. What happens if a person does cross into that? Stretch stops. That's it. And so it's very fast. And we spend a lot of time ensuring that that's what happens. And one other thing about safety is that we've mentioned OK, we're in the US and the EU. The EU has really stringent and a high bar for safety. So we're deployed there. That declaration of conformity we've gotten. So we take that same safety approach that meets that high level of safety required in Europe, and we use it across our whole fleet. So that's what we're using in the US. Also what we're using in Canada. So that high bar, we maintain it. Because we really are a safety first company when it comes to our robots at Boston Dynamics. You're interested in automation. You made the decision to go over. You've gotten deployed. You keep your people safe. How do you maintain your robot and protect your investment? Yeah. Great question because that's super important that you have to be able to rely on the robot. And to do that you have to ensure that it's well maintained. When it comes to Stretch, we really designed it for very little maintenance. Stretch is built for the rigors of the warehouse, right. So we know we're going to encounter dust. We know that boxes might fall upon Stretch. We know that it has to drive over bumps and stuff like that. So Stretch is a super robust product. To help with maintenance, I know we have Stretch Care. Can you tell us more about that. Yeah. Stretch Care is our program to ensure you have the uptime you need. We have service hubs all across the US and the EU to ensure, hey, any downtime is quickly addressed. We have that personnel. We have those depots across these main regions. That's important and it does set us apart from our competition. The next thing is, that I want to mention is that yes, we have people who can come on site, but we also have folks internally, technical service engineers, who are there to respond to questions. And so if there's any downtime, they can quickly respond, hey, try this troubleshooting step so they're ready to answer phones, ready to answer emails at all times. And we also have our service hub where folks can sign on to and get documentation and to learn, hey, maybe I should try this. Maybe I should try that. So we spend a lot of time ensuring that our customers are well cared for and well looked after. Well, while we've been talking, people have been submitting questions. Some of them have been submitted in advance. So we're going to answer a few of those. We're going to take a 1-minute break and get right back. Don't leave because we're going to answer one of the questions that we get the most, which is, what is the Roi of the robot. So we'll be right back. [MUSIC PLAYING] It is a blessing to have automation that you don't have to bring as many people in on the weekends, allowing them to spend more time with their family and their loved ones. So the predictability, especially on weekends, has been a really big benefit for us as far as being able to spread out our volume, and use the machinery to get a better Roi. You use it five days a week, one shift, it's hard. We're using it three shifts, five days a week. And sometimes day six and seven. So it's been a big help in predictability and scheduling staff and planning. So we're excited about the future. But we see other uses in the years to come and the partnership should continue to grow. [MUSIC PLAYING] Welcome back. Now we're going to take some of your questions. I've been looking at some of them that were submitted. And as I said, Grant, we've gotten quite a few questions on Roi. Can you tell us when is the investment of the robot realized, and how are customers measuring success? Yeah, the Roi Shirley, is two to three years. Of course, that's based on utilization and volume. But across the two shift operation customers today, that's what they're seeing, that two to three year Roi. So what it means is that if you are moving a million cases, or roughly 1,000 containers per year at a facility, Stretch is a no brainer. We're also getting questions around what kinds of freight can be moved with these solutions. Tell me about Stretch in particular. What can Stretch work with, and what can it not work with? We've talked about how we've moved 25 million cases. We talked about the range of material we're moving. So we can move a lot of different things. So that ranges from roughly 6 inch dimensions all the way up to 36 inch dimensions. So that's a wide range. Up to 50 pounds, we can move. When folks put Stretch in front of a container, they don't know what's inside that container. But Stretch just works its way through. So those cases can be heavy. These cases are oftentimes coming from across a bumpy road halfway across the world, then on a ship, and then across another bumpy road. So by the time they start unloading them, Stretch has to deal with boxes that are not in perfect shape, a lot of dust, and Stretch manages through that. So not only do we have a wide range of sizes and types of material we can handle, but it's also the conditions that they're in. Because real world conditions mean that you have to handle things that are not perfect. And when you talk about conditions, we get asked about what if boxes are crushed a bit. What if they get wet. What if they get soaked through. How does Stretch handle those? Of course, there's a range of options here. When they're crushed, Stretch will go after and try to get it onto the conveyor. Stretch has this smart gripper that can adjust to places where OK, a box might be deformed a bit. It's able to press down and make sure it has a good grasp. With wet boxes, oftentimes Stretch will just say, hey, we need a person to come in here and remove these wet boxes. People do appreciate that, too, because they want to be able to come in and document when some material has been damaged along the way. So it's important for operations to be able to do that. Yeah. Instead of having the robot move some soaked cases along, you're going to want to inspect those anyway. Yeah, exactly. Or a destroyed case goes everywhere. Socks are all over the container floor. You don't want that. So we also get asked about how Stretch handles containers that are not stacked perfectly or very messy containers. Can you tell us about the complexity of freight that the robot can handle. Sure. Yeah. Stretch is handling some fairly complex freight boxes that are not stacked perfectly. Walls that have different size boxes in them. So Stretch is doing that today. So we're able to handle that. Boxes that are crushed, boxes that are wedged in between each other. We're doing that today. And we're also spending a lot of time delivering better capabilities when it comes to that. So even as they're getting more and more messy, you can imagine parcel trucks, things like that, we're striving to get there and we're building those capabilities day in and day out for Stretch. We don't move canoes, we don't move tires today, and we don't move polybags today. Particularly when it comes to polybags and really messy freight, things like we see at these parcel places like I mentioned, we're spending a lot of time focused in on that problem today. We've also gotten a fair number of questions on Atlas. Our production version of Atlas was released this year. There's a lot of excitement around that. People are curious, what about Atlas for the warehouse? Do you think that's something that people will be seeing soon? When it comes to the warehouse, Boston Dynamics sees lots of opportunity in the warehouse. We're learning more things about the warehouse every day. The impact that Boston Dynamics has on the warehouse won't be just Stretch, right. We expect that Atlas will be there. Atlas does have an advantage on dexterous manipulation. There are those sorts of operations in a warehouse. So we do see Atlas seeking to do that in the future. We're looking to better understand where those places are. People in the crowd if you have a very specific spot where you see Atlas might work where you didn't think Stretch would, please reach out and we'd love to talk to you. Do you want to talk about some of the jobs that Atlas might do because it's got more of that dexterity, as opposed to Stretch with the powerful gripper? Yeah, it's a good question. I think the one that jumps out is each picking. Right so, Stretch does a good job in showing that cases get from inside the container into the warehouse. And oftentimes what folks do then is they open up the boxes and they decant all the material into another place so that way they can be sent out to stores or directly to customers. That sort of decanting, each picking, or when you're building an order, that sort of operation would be great for Atlas. Another question I hear a lot is how fast does the robot work, and does it work faster than a person? Yeah, good question. Look, there's a big range here. When you're moving 50 pounds boxes, you're probably moving at 500 cases per hour. But then the full range when you're using multi-pick, which we've talked about in the past, when you're using multi-pick you can go up to 800,000 900,000 CPH values that we're seeing across a full trailer. So it's a full range and it really comes down to the material you're moving. When it comes to comparing with a person, I think one of the big things there is that Stretch just works away. So when you press go on a container full of microwaves that are all 50 pounds, that speed is consistent. Folks, I've talked about how strenuous of a job it is, they'll start and they'll get tired along the way, understandably. So their rate will drift off. Stretch isn't in that way. So Stretch provides that consistent inbound of boxes that the rest of the warehouse machine can use and to plan for. Another thing we're getting asked is, do you offer RAS, robots assist service. Yes, we offer that. We seek to make it as seamless as possible for people to get started with Stretch. So we offer that. So if that's something important to you, please reach out. We'd love to talk to you about it. This is a question just for you, Grant. What is your favorite thing about Stretch or working on Stretch? Of course, I have to say, the people. It's been a great time building Stretch from a pilot machine to now, this industrial product that I said, when I buy clothes and put them under the tree for Christmas, I know some of them got moved by Stretch. And that journey has required people working really hard together and making hard choices together. And that camaraderie, and the really smart people I get to work with has just been a blast. You've been with Stretch for a long time, too. What is your favorite part about Stretch? Internally, I think one memory I have is one of the times we had a milestone celebration and we had this silly game of throwing boxes. And it was you and me. And I don't mean throwing boxes in the way that the industry means. I mean, literally us throwing boxes to see how far we could throw them. And I remember you beat me in that aspect. You mentioned that you love working with the people at Boston Dynamics. I do too, but I also really love meeting our customers. I really enjoyed getting on site, and seeing how the robot worked with different types of freight, different types of containers, that it would work harder to pull something out, that things were really jammed in there, out in the real world. And I also really loved meeting all the people, the associates, the project managers. I would ask and hear that people really did enjoy using our product. But for me, it was different to go there and to see it and to hear it from them and to meet them and develop those relationships. That's awesome. Yeah, they were great to work with and we've taken a lot of trips together. We really learn a lot from the people who are running our robots. And so it's fun to get out there and where they're saying, hey, we like this about Stretch. Oh, you need to change that about Stretch. That feedback is really fun. Well, thank you for joining us today. If you'd like more information, please reach out to us at bostondynamics.com. Thank you.