Transcription
What's up, guys? Wayne here. You're watching Equipment World. Welcome into another episode of The Dirt, the video podcast that brings you the latest in construction equipment, trucks, and construction industry trends. Coming up today on The Dirt, we're going to be talking about this guy, Spot, the robot dog created by Boston Dynamics. Now, by this point, you're probably familiar with, or you've at least seen a video of, Spot on the news or somewhere on social media. Or maybe you've seen video of this guy. This is Big Dog, Boston Dynamics' first robot that incited kind of a mix of laughter and fear when viral videos of it started rolling out more than a decade ago. And while Big Dog was the starting point, a more militarized, kind of carry heavy things version, Spot is really what Boston Dynamics has been after for a while now. A smaller, more agile robot that can be equipped for use in just about any industry. And in the last year or so, Boston Dynamics partnered with Trimble to figure out specific things, specific tasks that Spot could do on construction job sites. And because Spot is so small and versatile, it's the construction industry's first true roving robot. Kind of a big deal. So, with all of that in mind, today on The Dirt, we're going to be talking with Martin Holmgren. He is the General Manager for Buildings Field Solutions at Trimble. And we'll also be talking with Brian Ringley, the Construction Technology Manager at Boston Dynamics. So, with Martin and Brian on board, we're going to be talking about Spot, kind of the development that led to Spot, and what not only makes it such a good robot, but what about it makes it such a good fit for carrying out tasks in a construction environment. So, let's get into it.
Martin, Brian, thank you guys so much for joining us today on The Dirt. How are the two of you doing today?
Thanks, Wayne. Great to be here. Happy to be with you today.
Yeah, thanks for having us.
All right, so I, I kind of want to start at the beginning, the, the maybe the very beginning, just in terms of, uh, for, uh, the, the time, the timeline that Spot has been on. Brian, so for everyone out there who is somehow, uh, unfamiliar with Spot or the kind of, uh, the, the dog-like robotic line that Spot has kind of come from, give us a quick summary of what, you know, Spot is and, and what in general, uh, it is designed to do.
Yeah, so, you know, we like to call Spot an agile mobile robot, um, to differentiate its, uh, enhanced capabilities over other traditional forms of mobile robots. And really, it's the fact that it has legs. It's a quadruped or, or four-legged robot, which is why people hog. And those legs allow it to go up and over obstacles and allow it to traverse up and down stairs. Um, and these things are really important, uh, for the robot to be effective in our customer environments and things that are, are shortcomings of other types of terrestrial robots. And by virtue of that, we're providing a generic mobility solution for customers. If you can take automation anywhere in a human-purposed environment, that means that you can put valuable sensors on that robot and go anywhere in those environments to get data.
Yeah, I mean, that kind of answers my, the next kind of follow-up I had was, you know, uh, I've actually wondered this for years. Is, is it's not necessarily, I, you know, I'm wondering why a dog, why the form of a dog, uh, but it's, it's more so because of those, those four legs actually, uh, allow it to traverse more, more terrain, right? So it's not necessarily why a dog, but, but why four legs? Would that be the better way to ask that?
Yeah, I think that's right. Uh, you know, bipedal solutions, we, we have a bipedal humanoid robot called Atlas, but that's really more R&D. It's a very resource intensive to maintain the balance on on two legs. So, you know, it's, it's convergent, uh, evolution in a way. There are a lot of, you know, really effective animals on Earth that are able to traverse a lot of terrains with four legs, and we kind of arrived at the same solution with linked robotics. And it gives you both the dynamic mobility and the, the static stability required for customers to do the types of things they need to do with the robot.
So, is that kind of why, you know, in the, the early videos that, that we were seeing, gosh, I, I think the first videos came out back in 2008, um, but, uh, I was just doing some quick YouTube research, but like, is that why we saw so much, uh, kicking going on in those videos? Because it was all about trying to test, really, the resilience and the balance of these robots?
Yeah, yeah. And that's, that's from largely the era from 2004 to 2012, which, which kind of started with Big Dog and ended with LS3. We're developing larger quadrupeds, uh, to carry large amounts of weight, uh, over, over all terrains. And, yeah, I think it's partially, you know, to just show, to test, you know, does this thing work? Is it able to balance? But it's, it's about contending with the reality that when you put a robot into the real world, things that, things are going to happen to it that you can't plan for. So it has to be able to react to those things on the fly and make intelligent decisions. And a lot of them is just about, you know, basic balance.
Well, Martin, we all kind of have this, you know, I, I, it's, it's one of the, I, I'll never forget the first time that I kind of saw the, the video of Big Dog, uh, being tested, and it definitely had more of that kind of like military, uh, look to it, or, or maybe, you know, something out of some, uh, dystopian future movie. Martin, what was your, kind of, initial thoughts upon seeing this quadruped robotic technology that Boston Dynamics kind of pioneered for the first time?
So, I would actually like to take us all back another 10, 15 years. So, from a Trimble perspective, we've seen the opportunity of, of robotics and robotics in the construction industry for much longer. And we have worked with a variety of, of players, uh, academic and industrial, over the years. So, so we're absolutely committed and believe that robotics can provide productivity in the industry that, uh, doesn't exist today. But of course, what we know, the industry is way messier, way more challenging than what people think when they start to develop a robot for the building construction, of the construction industry. You know, it's, it's trucks hauling dirt, there are excavations going on. Um, if you're, if you're outdoors on a civil site, if you're indoors on the building side, there are ladders and building materials stacked up, and you have to navigate environments that just, every time we've done this before, we couldn't meaningfully achieve the productivity level that we needed. When I first saw Spot and its predecessors from Boston Dynamics, it, it, it so obviously presented an opportunity of, of carrying out work in what I would call almost an adverse environment. And a four-wheeled robot, uh, doesn't really do that. A track draw, but can't do that. You can't overcome these challenges easily. But, but Spot and the approach that Boston Dynamics team has taken, and especially when we started to see Spot and its productified version that we've seen here in the last few years, obviously we've been part of the program for a long time, working with Brian and his colleagues at Boston Dynamics. And the even the development that I have seen from when we started engaging with Spot to the performance and the behavior of Spot that we see today, it's, the, the exterior looks the same, but the way that it's improved over time is, it's really, really nice. And, and we are, we're now absolutely convinced, and that's why we announced this alliance between Boston Dynamics and Trimble, that we can take it to the next level and, and actually start to offer real productivity on this side, thanks to that four-legged quadruped behavior and capabilities that Boston Dynamics has delivered and developed over many years.
Brian, Martin brings up a really, uh, good point there, um, with, uh, you know, pointing out the differences between wheels and, and legs. And I think if you go back to just like, you know, go back to the, The Jetsons or Short Circuit or, uh, you know, Lost in Space, all these kind of like classic, um, kind of imaginings of, of robotic assistance in, in the future, um, it's, they all have wheels. Uh, I, that just literally just occurred to me. And obviously wheels are easier to implement, but, but as Martin was saying, they, they just don't go as many places as legs. So obviously wheels are easier, legs are harder. So, you know, what was, Big Dog and kind of this implementation of, of four legs, which definitely makes the balance problem a little bit easier, what, to your mind, to the company's mind, was that really kind of like, the, the most successful or the earliest successful kind of, uh, example of, of a robot with actual legs that could actually stand up?
I mean, this was actually formed, was actually at MIT spin-out. Uh, there was something called the Leg Lab, which was founded by our now Chairman, Mark Raibert, in the early '80s. And then it moved to MIT. It was at Carnegie Mellon, and then it moved to MIT in the late '80s, early '90s. And if you read Mark's early work, it's about how only about, you know, roughly half of the Earth's land mass is accessible by wheeled vehicles. Um, and, you know, he kind of envisioned a future in which robotics and automation could be useful, you know, across the entirety of the Earth's land mass. And, and legs seemed to be a way to do that because obviously, you know, we've seen a lot of animals who are able to do that with their legs. But I think what that speaks to, kind of in our mission as well, to make robots that are valuable for customers in these industries, is that we work with a lot of industries that have not necessarily been designed to accommodate automation as it stands today. So our construction site is not the same as say, a warehouse or a factory where it can accommodate, you know, fixed and wheeled automation. Um, our customers need automation that just works in their environments, that brings value to human-purposed environments without the need for them to invest additional resources and redesigning those environments, because that introduces friction into adoption. Um, and we can't have that. So I really think that from those earliest writings and thinking about the fact that, you know, paved roads had to be made to accommodate cars, that extends, kind of into the position we're at today, where we can just add value to these environments as they exist.
Kind of getting back to, um, you know, Spot and, and, and let's just kind of, I'll draw a straight line between, you know, obviously there's been different versions of this technology in between, but I'll draw a line between Big Dog and, and Spot. I feel like they're the most, um, maybe widely known, uh, of, of the kind of like process that, that this whole kind of thing has been. So how is Spot different today? You know, obviously he's smaller, he's a little bit more dog-like because he is smaller, but how is Spot different today than he was back when we were seeing those early videos 10, 12 years ago of Big Dog?
I mean, there, there are quite a few differences, as you can imagine. Um, you know, there are some big ones that have contributed to both, it's kind of appropriateness for our customer environments, as well as cost-effectiveness in those environments. So it's, it's all electric motors, for example, powered by lithium-ion battery, whereas earlier robots were often, you know, diesel and gasoline powered, um, or had hydraulic actuation versus electric actuation. The smaller form factor is good because you need the robot to be able to fit through standard-sized thresholds, move up and down standard-sized stairs, and, you know, even try to get into some smaller spaces that, that might actually be dangerous for a human worker to get into. Um, but it also creates a form factor that's easier to mass-manufacture. The more robots we sell, the more we manufacture, and the more the price can keep coming down, which in turn can, can make it affordable to even more customers. So that's something that's, that's very near and dear to us. Um, and then I think the biggest thing is, uh, software. You know, there are a lot of amazing robots out there in the world, but if you put them in the hands of an end user, you know, there really isn't anything the end user can do with that robot because they're not roboticists. So we recognize that our, the, that our customers are, you know, construction professionals, energy professionals, manufacturers, they're not roboticists, and they shouldn't need to be to get the value from Spot. So we build, you know, what I believe is really good and intuitive software to use that. And that's built off of our own SDK, as well, which allows Trimble software developers, for instance, to incorporate that robotic functionality into their own really great, uh, software that their customers already love and use.
And what I love about that is again, we talk about reducing friction for adoption. Is if I've been working in the field for years with with Trimble software and equipment, and now essentially what you're doing is just adding robotic control to that. You know, that's fantastic because then, to me as a worker, this is another useful tool that adds value to me. I don't even have to think of this as a complicated robot. You know, this is just another tool in my, in my kind of arsenal of useful scanners and surveying tools that allows me to get my job done more effectively.
Yeah, um, and Martin, kind of following up on what Brian just said, I mean, we kind of talked about two things there. We talked about size and we talked about improvements in software. Um, you know, my first question to you, Martin, on that, do you feel like getting to this size and, and this form was kind of essential, um, to, for, for this robot to actually perform construction work? So, so what are your thoughts on the size? And then, kind of the second question is, um, from Trimble's perspective, you know, you guys obviously have hardware that you can outfit this thing with. It, it can, it can carry hardware, it can carry sensors around. But talk about also the importance of what, of what Brian was talking about in terms of the software. So, do you think that those are kind of equal? Is one more important than the other? How do you guys see that?
I'll start by answering your question about this size. So, um, when it comes to size, much of the TR, the Trimble technology equipment that we would want to equip Spot with, if of a size that's very appropriate for Spot as it's designed today. If it had been, you know, 30% bigger or 30% smaller, I still think that that being a perfectly adequate solution for us. So size in itself, it is not some, not a critical absolute, but, but something that now is, is adequate. And I would say that the, the, uh, ability to traverse adverse, uh, terrain is, is now adequate. And that means that that we can start to, to adopt and deploy, uh, Spot as the technology into some, to some extent. I mean, when, with your background and your listeners' background, I think there is lots of analogies between what we think we can do with Spot and what we can do with an excavator, a dozer, a grader. It, it becomes a platform that we can equip with technology to make it tremendously more productive than, uh, it would have been before. Um, so that, so that's that's kind of on the, on the hardware side of things. If you talk about software, so obviously, um, the vast majority of the value comes from the combination of software and hardware, of really enabling specific workflows, whether that's grading or or moving dirt with a dozer or collecting information or indicating positions with a Spot robot. It really isn't hardware or software. It is the value coming from the combination of the two that, that I think is what brings a return on investment for for customers. And, but people buying these solutions now.
One thing that we, that's come up a couple of times now is, is autonomy. Martin just kind of alluded to it there. And actually, in our conversation that we were having before we actually started, uh, actually actually recording, what we were talking about, um, you know, we were kind of talking a little bit about, uh, automation and machine control and, and, and how, you know, robotics will kind of ultimately kind of just kind of slot into those type of workflows. But, but, but Brian, give us a little bit of an idea. I know that this thing can be controlled remotely, but we've also, you know, it's come up a couple of times, the, it's that it, that it can act autonomously. Kind of, kind of let us know, so it, it can rove autonomously, it can be, real quick, is it both, or is it one of the other? Kind of explain that interplay.
Yeah, the robot can be operated in person, it can be operated remotely, and it can be operated autonomously. And you can mix and match those modes of operation. So, for example, um, I had an LTE-enabled Spot that I was operating in the office when I kind of accessed the office during COVID. And I could, you know, send it on a mission where I wanted to go somewhere specific in that moment, using, you know, human intelligence and decision-making. Or I could record a path and sequence of actions that I wanted to be repeated autonomously. And the way that that works in our tablet software is that you pilot or drive the robot to the locations you want to go. You perform the actions you need to perform at those locations, whether it's taking a 360 image or performing a laser scan. All of that's saved into a mission file. And that mission file can be repeated. And that's when the robot is autonomous. So it's repeating those same locations and actions while also exhibiting safe obstacle avoidance. So it's stepping up and over obstacles that are, that are 30 centimeters or lower, stepping around obstacles that are higher than that, so that it can kind of complete the job and get that done. So you get the repeatability and the flexibility. There are even some features in the software that can send, you know, messages to the operator if there's something that's extremely large and they want the operator's help moving around it, and then it will resume autonomous behavior. Or there's a data capture point, again, due to the reality of the job sites, where that location is blocked, you know, say by a pallet of materials that's been placed in the way. So you can give the app, the robot the option to capture that data at that location because maybe some data is better than no data. Or you can tell the robot to proceed to the next waypoint and just bypass that. So we, we recognize that the customer needs quite a bit of flexibility. Yes, we want repeatability, yes, we want autonomy, but, you know, the fact of a construction site is you can't really predict what's going to be accessible on a given day. I think we've done a pretty good job of accommodating that. Now, autonomy for interior construction environments, you know, common commercial construction environments, is really using, you know, SLAM technologies that a lot of people are familiar with. It's kind of operating off of feature recognition to do its localization and navigation. One of the really cool things about working with Trimble is, um, they're really good at GPS. So, uh, you know, they were able to integrate their SPS 986 GNSS smart antenna. And, you know, I had the, the privilege of traveling to a customer site the other day where, you know, we were able to initiate autonomous missions simply by entering GPS coordinates. So now we're talking, we're not even walking the robot, we're not even recording anything anymore. We're getting autonomy, you know, in, in the five minutes it takes for me to locate the areas I want to go to and essentially list them out for the software. Um, and that customer, while they had people on site supervising Spot, you know, that robot was being sent its autonomous mission commands, uh, from Minnesota to Texas using public LTE. So, you know, that just shows again, that combination of remote operation and autonomous operation that makes this this so valuable. And it also shows that the autonomy isn't limited to what our engineers are providing in the top, the tablet software. It can actually be expanded upon and improved upon by our partners, like what we're seeing with, um, the Trimble SPS 986.
Kind of hearing you talk about just kind of how the interaction with Spot works in terms of signaling it to do some kind of, uh, command autonomously or, or connecting to do some kind of remote control like you were describing, um, during, during lockdown when you couldn't access the office. The thing that that brings to mind is just kind of the state of battery technology. You mentioned that everything on Spot is electric drive and lithium-ion powered. So how does Spot kind of make sure that it is, it is, it's, it's battery is full so that it can kind of receive these commands? Like, what is the charging method? How long does that take? Does it kind of manage that on its own? What does that look like?
Yeah, so, um, and to be completely transparent, that's really been one of the, one of the main limitations on autonomy thus far. The user does have to intervene at that 90 minutes to swap out a battery. So it's a 90-minute life on average for each of those batteries. We provide the customer with multiple batteries to swap that out. Um, those batteries can also be hot-swapped, meaning that you can plug the robot into our shore power unit and you can switch out those batteries without powering the robot down and interrupting, you know, the course of work that it's currently performing. But, um, we're really excited that, uh, we actually have our self-charging dock on a number of customer sites. It's prototype form, doing those, those tests. And that, that will be commercially available in 2021. And I mean, I think this is a big part of the shared vision also between Trimble and Boston Dynamics, is that that will no longer be a limitation on the autonomy. And it's not just that you can run arbitrarily long missions, that's obviously very valuable. It means that you have a robot that's always on, that's always waiting, uh, and available for you to perform work with it. And that's, that's really critical moving forward.
And, and Martin, kind of time to kind of talk a little bit about, um, those, those limitations that, you know, uh, that battery obviously plays on any kind of technology. It doesn't matter if it's robotics or drones or cars and trucks, whatever. But talk about, um, you know, those current limitations with that 90-minute battery life and, and how something like a self-charging, uh, kind of pad will, kind of change things or, or the way that you interact with the robot on a construction job site, specifically where we identified exactly the same shortcoming. The, the 90 minutes and having to have a person replacing the battery was, was definitely a showstopper for us. So when, when Brian and the team behind him at Boston Dynamics indicated that they were well on their way to the self-charging, uh, situation, is, is one of the triggers for us to engage and drive, uh, the agreement, uh, around the alliance, which we have together for the construction industry. But in, in specific, we see that many of the workflows that comes to mind first with our customers are workflows that requires the device to operate autonomously and often on hours when the rest of the labor force isn't there. So whether it is a, um, construction site, say a small to mid-sized building construction site where the dirt moving is, is, is in process, and of course, in, in our vision, we want as many of the heavy machines as possible to be equipment machine control, which equipped with Trimble GPS, uh, or, or laser sensors, uh, so that we would collect much of the information around the site and the, uh, dirt that's been moved from one day to another without, uh, the Spot there. There are, as we, as we all know, uh, part of the fleet that is not equipped with machine control, haul trucks, dirt being being moved, or, or new equipment materials coming into a site. So what we envision is that by the end of the day, you tell Spot, hey, when I arrive next morning, I want to know the digital terrain model for, for these areas on the site. And then, then you leave, and Spot goes out, uses 80 minutes to, to measure part of that location, goes back and charges, and, and then continues on with its operation, essentially all through the, the night. And, by when, when you're back in the morning, you, uh, have a complete model of all the parts of the site that isn't automatically measured with the heavy equipment.
Yeah, I mean, telling it to go charge is definitely a, a lot easier of a process than actually performing those hot swaps, uh, uh, you know, by hand, for sure. Um, well, uh, you know, I, I, I, I kind of want to get into a little bit more, uh, about Spot's kind of, uh, uh, you know, unique kind of fit in the construction world, and, and how it, kind of, how this became an industry that became kind of top of mind, um, you know, for the folks at Boston Dynamics. Uh, uh, you know, we've talked a little bit about, Martin, where, you know, you guys have been kind of excited about, at Trimble, have been kind of excited about the potential of this technology for a while. But, yeah, there's plenty of folks in the industry, you know, who probably look at this thing and think, you know, a, a robot dog, uh, you know, how, how is that, how is that gonna help me? Like, you know, because I think most people when they think about robots, when they think about construction, and they think about things that can help on the job site, I think that the, I think that the, uh, knee-jerk reaction is probably going to be, you know, can it carry stuff around? Can it, can it move dirt? Can it, uh, you know, is it gonna operate equipment? You know, you know, things like that, rather than kind of the more, the more passive things, or the more time-consuming things, like surveying, like, uh, like job site progress photos, like safety monitoring, stuff like that. Um, you know, Brian, construction is obviously in your actual job title, so it's a clear focus for Boston Dynamics as a company now. This isn't just kind of like, uh, you know, a partnership they've undertaken to kind of get into one additional industry. This is, this is your focus. So, you know, was construction always kind of back of mind as this thing was being developed from, from Big Dog onward, um, in terms of like a potential application for this device? Um, so, yeah, give us, give us an overview of how and why this, this industry was seen as a really good fit for Spot within Boston Dynamics.
I mean, I think that when, when Boston Dynamics transitioned from Google to SoftBank, it was really about saying, hey, we've done a lot of research into bringing these legged robots into kind of more domestic and industrial human-purposed environments. Now, can we find a product fit? Can we find use cases that have ROI on them? Uh, talk to a lot of different industries and really learn from this potential customer base. But the story's maybe just better told through my own experience because I started as a customer. In my prior job, I was a construction automation researcher, and, you know, had been focused on a lot of kind of industrialized construction techniques with traditional industrial automation. But then, as, as our company, um, you know, was really looking at ways to be more efficient on job sites, we realized that there was, there was a huge data gap. We realized that, you know, in order to solve a lot of our business problems, we, we kind of needed this, this data infrastructure and better job site tracking across our entire fleet. Um, so we had, we had done some experiments with, you know, people living around and doing that data capture, and had found that, you know, hey, they found it to be dull and, and low-skilled work that, you know, they had other more urgent pressing things they wanted to deal with, communication with other job site stakeholders, the staging of the site for the next day, you know, getting home to their families on time, for example. Um, and, you know, they were really, you know, the superintendents and assistant superintendents of our job sites were really championing the automation of data capture. So when we started using robotics, um, you know, there was some really great robotic technologies we were testing, and, you know, that integrated with some of our data capture software, but they were all wheeled or treaded, and they all failed to get everywhere we needed them to get. In 2017, if you had asked me whether Spot would be the answer for this, I, I would have thought that it was overbuilt for the task. But, you know, truth be told, there also wasn't just a lot of knowledge in general industry about both capabilities and cost of this technology. So by the time 2018 came around and Boston Dynamics had released the, the initial construction videos of the alpha Spots on some of their Japanese customers' construction sites, I immediately, you know, made an appointment with my now boss, Michael Perry. Our company drove up to Boston, um, and we started bringing it onto our sites and making it work. And in the first week, it proved it could go anywhere on their job site, it could do it autonomously, and it could, all we could also use the SDK to upload the data it collected to our internal software system. So we had essentially checked all three boxes immediately, and I really knew that this was the future of the construction industry. And yes, we do get a lot of customers interested in, you know, well, you know, when I think of robots, I think of of earth moving and I think of actual, you know, construction labor tasks on site. And I think there's a future there for both Spot and other companion robots on the job site. But again, going back to this notion of a data infrastructure, you really have to have an excellent reality capture program. You have to have really precise surveying of your job sites, and you have to understand how things are changing from day to day for a lot of those other things to be automated in the first place. This is really setting the foundations for a future where there are lots of different types of robots automating lots of different types on site and augmenting the workforce in an intelligent way. I like what Martin said, which is, you know, it's not really about, you know, there's not like lights-out construction where all the human workers go away and it's fully automated. It's about augmenting them by, you know, running a night shift, essentially staging the site, scanning the site, providing updates and reports to all of their software systems so that when the crew comes in in the morning, you know, the job plan's laid out and you're getting the best of both worlds, both human intelligence and, and machine automation.
So, I, I Brian, you're absolutely right. Uh, the, the only other aspect that I would want to emphasize is dangerous locations. So we talk about autonomy, and autonomy is very important for our market success forward. Uh, but even the assistive versions where, where Spot is controlled by a human. So Brian decided that you were down in, in visiting in Texas just recently. I, I do know that there was, uh, infestation of water moccasins. That's right. On this, on the site. So that, to me, that'd be one obvious, uh, use case, so to speak. And, and today, I've been actually approached by, by two different customers, one in the US and one in Europe, that has large-scale tunnel applications where, uh, the tunnels have been under circumstances that, uh, they first want to scan and then send in humans into the site. And, and they both approached me to discuss, hey, how could we equip, uh, Spot with a laser scanner from Trimble? And even if it's all manually operated from someone standing in the tunnel opening, that has a tremendous value for these customers.
Yes, I mean, we, uh, that actually is a really nice segue because, uh, I mean, we, we just kind of talked about some of the, some of the, the ways that it can actually make work safer, um, by, by going into some kind of dangerous areas that you wouldn't want, uh, a person to go into, whether it be a water moccasin infestation or, you know, a tunnel under construction, a giant hole in the ground. But, you know, what, you know, this is a question for both of you, so I'll, so, you know, each of you can, uh, can take a stab at this, but, um, you know, what, what specifically, what are some of the tasks apart from what, uh, Martin was just talking about? What are some specific tasks that that Spot can perform right now, today, that will immediately improve the lives of workers? We were talking about some of the mundane stuff like data collection, we've talked about safety, but what are, what are some of the other kind of real-world, uh, things? And what are some of the equipment that might be necessary for Spot to be able to, uh, pull those things off?
You know, we're, I think Mark made a great point about, you know, going into any potentially hazardous environment, right? Um, you know, enclosed spaces on construction sites. You know, we've also got a lot of the energy customers, so you can imagine radiated environments, environments that haven't been de-energized but require inspection. I mean, they're huge ROI opportunities for those. I mean, you really can't put a price on that kind of safety. Um, and also, you know, as we moved through this era of COVID, we've also kind of been faced with the fact that every environment is unfortunately potentially hazardous to human health. So, you know, we feel we've been able to help contribute to that. And, you know, as job sites have kept running, but minimizing volume on site, we've been able to augment that workforce to an extent, kind of help out that scenario. So I think Martin's right to point that out, and that is that is the most valuable reason to put robots on these sites is to protect your workforce. Um, you know, on the data capture side, we've talked about 360 images and videos. I mean, it's a very common way to track progress on job sites. And, you know, it can be anything as sophisticated as as our kind of Spot Cam PTZ, which allows for pan, tilt, zoom, and kind of zooming in on objects of interest and having it be able to zoom in on those kind of repetitively and honestly in future missions as well to track that item. But it can also be something like a Rico Theta, right? It can be a handheld, you know, low-cost 360 camera that, you know, people on the job site already know how to use, how to use, but just want to be able to on and add some mobility to that sensor. So there's really quite a bit of flexibility there. And then, you know, the, the other, the thing that customers are just hugely excited about is the automation of laser scanning. Um, you know, I, I myself have worked on, on and with reality capture teams in the past, and, you know, as you're moving tripod from, from spot to spot, and you're doing those high-resolution scans that just take forever, and, you know, you kind of hide behind something so you're not captured in the scan, maybe play a video game on your phone while you wait. It's really, it's just not a great use of a skilled professional's time. It's important work, but, you know, those professionals can be freed up to be doing higher value things. So, you know, people are really intrigued by by this. And, you know, I, I love using the X7 on Spot. It's a really, it's an incredible, it's night and day, honestly, like I get pretty excited about just because I have done this work manually, and the ability to have an autonomous mission that Joe goes out and captures my data while I work on other things, you know, the robot returns to me, and I'm able to just download that data straight to the, you know, Trimble X7 tablet, perform my, you know, automatic registration on the tablet, so I'm still in the field, right? Like, and I've done everything. So by the time I get back to the office, I've got a fully registered point cloud that I can import, you know, into BIM or wherever it needs to go for me to do valuable work. So it's, it's really something special. I'm really excited about it, and it definitely captures the attention of of customers. You know, and you, beyond the kind of imaging sensors, we've seen a lot of other creative uses in the industry as well. Um, environmental sensors, you know, using, using the dog to sniff, so to speak, for, for potentially dangerous chemicals, silica is a big one on job sites. Um, you know, again, to maintain the respiratory health of workers, and the list kind of goes on and on. And I think what we'll be looking for in 2021 is as we move from sensing to manipulation. So we will have the Spot Arm available, which, you know, allows the robot to kind of grasp and carry objects on the job site. And, you know, it's early days for that. We know that it enhances autonomy by doing things like opening doors, you know, potentially dragging small obstacles out of the way, things of that nature. But we're excited to see how that brings value to our customers.
I, if I would add some specific, uh, civil construction or heavy, heavy equipment aspects of this, I would say, hey, there's a lot of environments, uh, around the world where construction and explosives goes together. Uh, that's another example of those dangerous, potentially dangerous environments that we talked about where, where you may not always want people to be present, uh, at all times. While, uh, Spot, with Spot, that's much easier, of course. Um, I would also say, if we talk about specific workflows, guiding on the construction site to specific locations, if we take the, again, Brian mentioned the integration of GNSS, Trimble GPS technology, and Spot, we now essentially have a Spot that can find its way on its own. So if you have a delivery truck of some sort delivering a specific set of materials, you can send Spot off as the guide to make sure that whatever is delivered on the site is deposited on the right spot. So that's, that's just another example of the workflow where these dull, repetitive activities that that brings, you know, little real value, but still has to be done. I think we all know that the industry is struggling with access to a workforce, with a workforce that's sometimes aging, um, and, and exiting out over the coming five, ten years to, to a great extent. And having robots being able to step in and have people focus on the high-value aspects of all these workflows, I think is, is critical to our thinking of how, uh, Spot will really become a mainstream part of the construction in the coming years.
It's another good segue, Martin, uh, for, for this next question. But, yeah, like, so obviously, um, this, uh, even, even though Spot is really kind of, um, to me, it's, it's, it's, it's a, it's a full realization of, kind of, the work that was being done initially with, with Big Dog, and, and, you know, it's, it's a more, you know, I think obviously looking at, you know, those, those initial videos of Big Dog, I think a lot of the kind of, uh, maybe, uh, like, kind of bewilderment, and some of the reasons that those videos went viral, um, is because like people just didn't, couldn't look at it and immediately know, like, exactly what it was for. I think that you can look at at Spot, though, and, you know, if he's, if he's, if he's carrying a sensor array or, uh, you know, a, some kind of, you know, he's been turned into a GPS rover, you know, you, he's got a 360 camera on, on his back, you immediately kind of like, oh, okay, I get it. Like, this, this thing can, you know, this, this thing can carry out tasks. There's a bunch of industries that can be used for. It's just, to me, this, this version of the robot, of the, uh, you know, this, this, the way that the platform has advanced, just a lot more tangible at a glance, that, that of the work and the kind of usefulness of it. But obviously, it's not, it's, this isn't the, you know, fully realized vision. This, this, this, this thing is still not perfect yet. Like any other kind of technology, it's only going to get better. So, kind of the, the, the next to last question that I have for, for the two of you is, you know, what are you hoping that Spot, you know, what can improve on Spot that will enable it, you know, for, for it, you, you, you know, we'll enable it to do what you both want it to be able to do in the future? And what specifically, uh, Martin, would you like to see it be able to do? And, and Brian, what specifically would you like to see it be able to do? Because you both obviously have, you know, the very construction-heavy backgrounds. So, I'm imagining that there's probably some kind of specific things that, you know, each of you would like to see. So, uh, whichever, whoever wants to go first, but yeah, what needs to get better and, and what are you hoping that it'll be able to do in the future?
If I could, if I can take a step at that first, so Trimble, of course, what we bring to the party is a situational, situational awareness. So with the GPS sensors, with the laser sensors, we always have a good understanding of the surroundings. So when you drop Spot in a new place today, it has to form an understanding of the environment in, in an evolutionary way. So one of the things that we are looking to immediately add to Spot's capabilities is essentially the ability to take all that data, whether it comes from GPS, whether it comes from laser scanners, whether it comes from a, a BIM model for a bridge or a BIM model for a, a building. We want to pull all that information together and essentially give it to Spot to enable it to function in its environment in a much more efficient way than it does today. So if you, if you put a Spot that isn't Trimble-enabled in front of an obstacle that it hasn't really wrapped its head around, you will think, hey, this is a stupid robot because it will try, you know, approach it from several different angles because it doesn't have the situational awareness. So that's, that's one where, where I think the amazing abilities of the four-legged robots combined with a greater situational awareness will, will make a significantly higher value for the users of these robots.
I totally agree with that. You know, I, there are so many, I mean, I'm a product manager, so everything I want to be improved, um, the usability of the software and the tablet, again, I think is really key, as we essentially, as we move out of innovation groups and into operationalizing this across companies, you know, it just has to be intuitive for your end users, and it's really about design. There are behavior improvements that we are always making. You know, the robot is always going to follow us. When the robot falls, it's going to fall more elegantly, right? Where there are always things to kind of shave off there. You know, we've gone from just doing standard stairs to riserless stairs, to riserless and graded treaded stairs. We're getting more flexibility there. We'd have initial success on even steeper stairs, like stair towers and scaffolding. Um, you know, tight spaces, again, we talked about tight enclosed spaces that could be dangerous. We're finding ways, you know, to shimmy Spot into smaller and smaller areas, you know, while still understanding obstacle avoidance. And then, yeah, I mean, Martin touched on, you know, how do we, how do we use the sensors of our customers, right? This is part of the value of these partnerships to build more semantic understanding of the environment. You know, sense there are things where, you know, Spot, Spot doesn't know me from a column, right? It's just avoiding it. Just
doesn't want to run into something. That's how it's safe, but there's a lot of value in it having mission of what it is. It is trying to avoid or maneuver around.
And then there are external things that I think, you know, I'd like to see improve, uh, just for all of robotics' sake. I mean, I think better connected job sites. You know, in the construction industry, it's always going to be a struggle to have the right network infrastructures because these are ephemeral or temporary environments. You don't want to invest a lot in infrastructure, but we've seen some promising things with, you know, anything from internet control to control over public LTE as well as private LTE spectrums. 5G is obviously very promising because it's going to give us more bandwidth to to do more things with a robot, to send more data. Um, and, you know, like I said, the list really goes goes on and on. Um, but, you know, that the fun thing about working with hardware and, you know, I'm sure Mark can relate to this, is that you can just keep milking improvements out of your hardware with with software improvements over time. And there will be hardware improvements too, you know, as introductions of things like the dock and the arm. Um, but, you know, there's so much more that the current version of this robot can do and will be able to do, you know, even just within the coming year.
Well, I think another natural question, and this is kind of where we'll where we'll end things, but, um, you know, I think another natural question after after hearing all about this, um, is going to be price. Um, so, you know, over the course of the conversation, I I kept being reminded by just kind of by some of the functionality of of drones and kind of the benefits that drones have brought to the construction site. You know, obviously from an imaging perspective, a surveying perspective, and clearly Spot, you know, a big advantage is that, you know, while a drone's, you know, you could put certain sensors on drones, um, but but really the drone's biggest kind of uh benefit is going to be from an imaging perspective. Clearly, there is there's a lot of potential for Spot because it is on the ground, because it can move closer to those objects, because it it, you know, uh, it it is just a little bit more capable in in in that way. I think that there's even more applications that that it's got kind of up its sleeve as we kind of move forward. And I draw that kind of that line to drones because I was actually, you know, back in July whenever we first got, I think the Spot Explorer, um, went on sale, uh, earlier this year for, uh, I think the the price tag was about $74,500. You know, I that's actually not, you know, when you start talking about, um, industrial drones, it's actually pretty close to an industrial drone, um, especially when you, you know, the ones that are really serious about surveying, uh, and stuff like that. Maybe maybe a little bit more expensive, but but actually pretty close.
So, you know, I think my overall, uh, kind of question is, you know, that $74,500 price tag for the Explorer, you know, how does that relate to even obviously I would imagine that you can you can spend more money on it if you want? How much does it depend on the equipment that you equip it with? How do people go and get it and see one in person? I I know that the Trimble dealer network is is a big part of this. So, yeah, how does that work in terms of trying to get an idea of of pricing and and where to actually uh get one?
Let's start with where to actually get get one. So Trimble will be distributing the the Trimble enabled Spot, the Trimble built out Spot version through our standard dealer network. So, uh, will be available in a variety of countries around the world as we gain compliance for more and more countries. We've announced the European Union, U.S. and Canada, Australia, New Zealand, Singapore, Korea, Japan, right now. And Trimble, of course, has hundreds of dealers in in these locations. Dealers that are, uh, in a uniquely advantageous position in that we deal with complex optomechanical, uh, equipment every day and have the cert the local service capabilities, the local support capabilities of and ensuring continuity of operation of customers in in a way that is unique, I would say, in the in the construction space. So, so yes, you will be able to get both by get support and get service directly from Trimble. Of course, if if the robot is is run over by a bulldozer, we will need to call the Boston Dynamics team in and and do more advanced surgery, but but, uh, yes, it's a Trimble responsibility and the Trimble interface to the, uh, construction customers.
Yeah, and I think, um, you know, something I'm really excited about is that, you know, when somebody comes to us, you know, there's a conversation. What's the application? What do you want? There's kind of an a la carte menu of additional payload options that can go on top of that base Spot. But what I like about this partnership is now you just go to your Trimble dealer and you say, I want a really effective autonomous scanning solution. Right? You're not worrying about bits and pieces that have to be assembled. That's not your responsibility. You're getting the robot, you're getting the mounting kit, you're getting the scanner, and you're getting the software that's, you know, compatible with with your current software subscription and current control device. And then what that means is the equipment arrives and you get to work. You know, there isn't this this phase of kind of assembly and research and discovery. You know, you just have a useful tool. And that's that's really been the goal. You know, through our early adopter program this year is to take this thing from being kind of something that's cool, which it is, it's like very cool and interesting technology, to something that is a piece of useful equipment on these sites.
Brian touched on it briefly, which is the early experience program. So, um, we announced jointly here just a few weeks ago, the the alliance that we are starting to work together, develop solution together. Uh, that press release stated that we will start shipping product in Q2 next year, 2021. Uh, what we're doing between now and then, and I'm being bombarded with questions about this, I I know that there's interest out there. It's what we call the early experience program. And and unfortunately, we have to be very selective here, but we we are engaging a small subset of customers in a very intimate process, which involves direct interactions with product managers on both sides, both on Dynamics and Trimble, in essentially guiding the combined set of companies in what is the most valuable solution we should go after first, evaluate beta versions of that solution, and, uh, to speak, help us get to the end goals. That's that's a program that we are running. Requires you to buy both Spot and the payloads from Trimble, but, uh, it's an interesting program that we are able to add a few more customers onto that program here in the coming months.
Yeah, and just to just to echo that, you know, as we work to polish this this integrated solution, the technology works together today. You know, I'm out here running X7 scans, but we mentioned being on a customer site testing the GNSS capabilities for the GPS autonomy. It works now. So, yeah, uh, we're really excited about, you know, getting to work kind of one-on-one with the right customers, um, and for them to have the opportunity to shape the product as well. You know, we we always stay humble about the things that we're good at, um, you know, on our side, which are robots, um, and and always always being open to learning from the customer and better understanding the industry. I was obviously hired because of my understanding of the industry, but there's always so much to learn. And, you know, unfortunately, during COVID, nothing beats being on the site. So, I think that I think that's one of the most challenging parts of COVID is is having to be creative about remote support. But, you know, there's no better way to do that than with with robotic technology. So, um, you know, we've we've got to to prove that out at least.
All right, guys, well, that's that's pretty much all the time we had. This was such an interesting, uh, discussion. I I could talk about this for for another hour, um, but this is all the time we had today. Uh, but, uh, but Martin, thank you so much for joining us. Brian, thank you so much for joining us. And, and both of you guys, uh, take care and, and I hope that we have more, uh, Spot news to discuss, uh, sometime here in the next year or so. Thank you very much, uh, Wayne, and I look forward to be on a site with Spot and you and Brian, and, uh, we're gonna have some fun together. Yeah, thank you. Thank you so much for taking the time. This is we it was really great to get, you know, a lot more in depth, um, into this Trimble of also Dynamics relationship.
All right, guys, that's gonna wrap up this episode of the dirt. But as always, we want to hear from you guys. Hop into the comments. Let us know what you think about this discussion. Let us know what you think about the future of robotics and construction, the capabilities of this robot, what types of tasks you would like to see Spot be able to carry out in the future. Give us your thoughts in the comments below. And if you like this video, found the information useful, interesting, whatever, please do us a favor and hit that like button below. It really helps our channel out. And if you want more coverage of the construction industry, construction technology, heavy equipment, trucks, and more, visit us at our website at equipmentworld.com. And while you're there, subscribe to our daily newsletter to stay on top of all the latest headlines. And if you want more videos on construction equipment in the construction industry, trucks, and more, you know what to do. If you haven't already, subscribe right here on YouTube and turn on notifications. Hit the bell so you're receiving up to the minute alerts whenever we drop a new video. Thank you guys so much for watching. We really appreciate your time and we'll see you next time.