From Legacy to Leading Edge: Automating with Mobile Robots

Hi, everyone. If you're just joining us, thanks for being here today for our webinar, "From Legacy to Leading Edge. I'm Makayla Krupp, and I'll be your moderator for today's session. Today, we're going to be talking about how to implement an effective robotics program at your facilities with Spot and Boston Dynamics. So let's kick things off with a short video. [VIDEO PLAYBACK] - We manufactured a future here at Michelin Tire. And I don't think you can get more into future than a mobile robot. [MUSIC PLAYING] My name is Wayne Pinder. I'm the tech methods and reliability manager here at US5 in Lexington, South Carolina. We're a passenger tire and light truck division of Michelin. - My name is Ryan Burns. I work in the reliability and methods department. We make sure that all the equipment is running at its optimal conditions within the plant. - We've been using predictive tools here for quite some time. We do all analysis, vibration analysis, infrared, ultrasound. - Getting ahead of equipment failures is important because it affects our production output. If we can predict a failure and we can plan and schedule the work to fix the issue before it becomes an unplanned breakdown, then we're able to increase our output as a company in a tire producer. - We were like kids at Christmas when we first got Spot, and we couldn't wait on the training. So we took it out of the box, powered it up, and started walking around. We went up and got our training. And then Ryan started programming the missions the week we got back. We've created 72 work orders to date in the system using Spot. - From the missions that we have set up, Spot's gone out and has detected a few thermal issues with motors that were missing some fans on them. And so we were able to identify that and get that corrected. And then also, we've detected 66 air leak work orders in the system, where we want to reduce our energy consumption. - We see Spot as the future. Hopefully, we have Spots in every North American facility. And a facility the size of US5, we probably need a whole dog pound or multiple Spots here to actually do what we need to do. - Moving forward with Spot, what we'd like to do is we want to improve on our missions that we've already got with the Boston Dynamics team, have them come in and set up a mission within our boiler house. And then from there, we'd like to continue the deployment process throughout the rest of the factory. Spot is a member of our maintenance team. The future is to have more Spots so that we can improve on our inspections and improve our overall output as a company here at US5. [END PLAYBACK] MAKAYLA KRUPP: Well, with that, we'll go ahead and get started with our content. Some quick housekeeping things before we get started. You can reach us in the chat if you're noticing troubleshooting issues like that. But if you have questions, please put them in the Q&A box. We'll have moderators monitoring that, and we're going to have a couple opportunities to answer those both in the middle and at the end of the webinar. So put your questions in the Q&A please. And we are recording. And this will be available as a recording after the webinar. So you can check that out when we're done today. With that, I'm going to hand it over to our presenters to introduce themselves. Robson, do you want to get us started? ROBSON ADEM: Yeah. Thanks, Makayla My name is Robson. I lead the field application engineering team here at Boston Dynamics. I've had the privilege of working with hundreds of customers deploying Spot out in the world. So I'm excited to talk about our experience in that with Adam. ADAM COMOLETTI: Yeah. And hi, everyone. I'm Adam Comoletti. I lead the sales engineering team here at Boston Dynamics. My team serves as the technical bridge between our customers and our engineering teams, focusing really on all things Spot during the pre-sale process. So as we jump in, I just want to take a moment to look at how far we've come in the world of robotics and automation and how it shaped the era we're living in today. So back in the 1990s, we started to see the rise of a lot of industrial automation, especially in automotive manufacturing. Robots were mainly doing tasks like welding, painting, and assembly. And these were repetitive jobs that benefited from speed and precision. And then by the 2000s, those capabilities started to expand into newer industries. We saw robotics move into things like logistics and warehousing, and the earliest version of collaborative robots began to appear. And then in the 2010s, collaborative robots really started to take off. And with the help of AI and machine learning, robots became smarter, more adaptable and better able to work alongside humans. And now, in the 2020s, where we are now, we've entered what we call the era of intelligent automation. So robots are becoming increasingly autonomous and using advanced AI to perform complex tasks with less human oversight. So we really are living in a different era, one where intelligent mobile robots are becoming everyday tools and industry. And that's exactly where Spot fits in. So before we even introduce Spot into the equation, let's take another step back and look at some of the very real challenges manufacturers are dealing with today. So legacy equipment, if your facility is like the most that I visit, a large portion of your equipment was installed before IoT was mainstream, and that means it just wasn't built with connectivity in mind. So retrofitting all that with fixed sensors, it can be very expensive and it's sometimes just not even practical. So second, getting real-time data. It's incredibly valuable for predictive maintenance, but monitoring everything continuously, especially manually, comes at a high cost in labor and in time. And then there's the workforce issue. Just labor shortages are hitting nearly every industry. And the last thing you want is to put skilled workers away from high-value tasks just so they can walk around and do inspections. And of course, safety is always top of mind. Some inspection points are tucked away in hard to reach areas or even hazardous areas. Sending in a person every day increases risk and liability. Just send in Spot. And this is exactly where robotic inspections and Spot come in. So Spot can take on those repeatable, time-consuming inspection tasks. It provides cost effective continuous monitoring without the need for retrofitting every piece of equipment. It captures consistent, standardized data that helps predict failures early and reduce risk, let your team focus on solving problems, not just finding them. And again, most importantly, Spot can safely go places where people just shouldn't. So let's take a look today at what Spot can actually do. These aren't like future features or concepts. These are real deployable tools that are being used right now in the field. So first, digital twin creation. So using the Leica BLK ARC payload, Spot can autonomously scan your facility to generate detailed 3D point clouds. And this gives you the ability to create a digital twin, a highly accurate spatial model of your environment that can be used for planning, monitoring, or even simulation. Visual AI. So Spot Camera payload, which is aptly named Spot Cam, can capture images of nearly anything like analog gauges, fluid levels, open doors, or even puddles on the ground. And using visual AI, we can interpret those images automatically, so you get insights like the gauge reads x. This door is open or I don't there's a leak in this area. Then we have thermal analysis. So with the thermal camera that's built into the Spot Cam, we can monitor asset temperature trends and catch anomalies before they become failures. So if a motor is running too hot or not hot enough, you can flag that immediately based on whatever thresholds you define. Now our acoustic sensor, so acoustic leak detection. So using our acoustic imaging payload, Spot can hear what humans can't. It can pinpoint compressed air or gas leaks by detecting ultrasonic frequencies that are otherwise impossible to catch. And it can do this even in loud environments, because it's only tuned to listen for that specific frequency. Now, using that same payload, we can also identify early signs of mechanical issues. So that payload will listen for changes in sound signatures like bearing wear or misalignment. And it will alert you before a minor vibration becomes a costly equipment failure. So I want to dive more into the AI visual inspections we've been talking about, or we like to call them AIVI So with AIVI, we're turning Spot Cameras into frontline sensors for safety, quality, and compliance and actively identifying issues that matter. So what you're seeing on the screen here, this is a real-world example of Spot detecting a puddle on the floor. So AIVI doesn't just capture the defined images. It recognizes the hazard, flags it, and sends an alert. So whether it's a spill, a gauge, or a door that's unexpectedly open, Spot can give you these insights. This also has huge implications for safety, catching things like spills, obstructions, or even missing PPE before someone walks into a dangerous situation. It also helps with routine operational checks. You don't need to send someone out on a time-consuming inspection just to see if a gauge is in range or if equipment looks off. Spot can do that for you around the clock. And compliance. So from a compliance standpoint, it helps you stay audit-ready. You get time stamped image-based records of the inspections. Spot runs with alerts for anything out of the ordinary. And this is really what we mean when we talk about intelligent automation. Not just collecting data, but turning that data into action. Now, I don't know if we're still having audio issues, but if we're not, how, about we hear about AIVI from [? Cargill ?] instead of just me. [VIDEO PLAYBACK] - One of the things we've been recently trialing out is AI Visual Inspections. So we're using the robot to walk around the plant, check things out, and then assess via the AI natural language questions, what it finds. - So you locate what you're trying to capture take a picture. And then at the end, you basically ask a question. And this question could be a numerical one, a yes or no, or even an open-ended one. [? Sideview ?] is a 360 plant view. What we're currently using it for is to support our food and safety quality team by basically having another set of eye in the plant. Spot walks this route multiple times a day, and then hes can identify leaks of seeds and oil early. And you can also historically look back. [END PLAYBACK] ADAM COMOLETTI: All right. So we've talked all about Spot's hardware, the sensors, and the AI. But how do you actually manage all of that? That's where our Orbit software comes in. So Orbit is our software platform that brings together everything Spot sees, hears, and does and turns it into actionable insights for your operations. So first, Orbit is going to give you real-time visibility into facility. You can monitor conditions live, get alerts as soon as issues are detected, make faster, smarter decisions before downtime happens. Now, second, Orbit makes it easy to analyze trends over time. You can track temperature drift, pressure changes, equipment, vibration, anything Spot is monitoring. And you can use that data to predict failures and optimize performance. Now, Orbit also helps you manage your entire fleet of robots. So whether it's scheduling inspection missions or checking the health of each unit or deploying updates at all happens from this one centralized platform. Now what if you need to take control manually, say you got an alert and you want to take the robot, drive it over and confirm the alert. Orbit supports remote teleoperation, so you can drive Spot from anywhere if you need eyes or ears on something quickly and safely. Data integration. So not all customers want all their data to live in Orbit. An Orbit lets you integrate all this data into your existing workflows. So whether it's a CMMS system, a maintenance dashboard, or any other reporting tool, Spot becomes part of your broader digital ecosystem. So with all that, you can see that Orbit is more than just a dashboard. It's the digital backbone that transforms the raw inspection data Spot collects into real operational intelligence. But with that, I'll turn it back over to Makayla so we can jump into our first Q&A. MAKAYLA KRUPP: Great. I see some coming in the chat here. So remember to keep putting your questions in the Q&A box and we'll have another chance to answer some more at the end. But we did have some pre-submitted questions. We had a couple about how Spot navigates stairs and small obstacles, or how it might handle if there were vehicle traffic or something passing in front of it. Robson, do you want to tell us a little bit about how Spot would handle those situations? ROBSON ADEM: Absolutely. So Spot is designed to perform robustly in many environments and terrains. In addition, it's also designed to handle stairs that are congruent with the OSHA stair spec. In addition to that, Spot is aware of its surroundings. It's aware of static obstacles, such as containers. In addition to that, it's also available-- it's also able to detect moving objects such as people and forklift. And then we also have baked in autonomy behaviors that are paired with audio and visual cues for humans that are working on Spot to be aware of it's next action. So those would be great resources. MAKAYLA KRUPP: We have a couple questions coming in live. Can faces or other objects be blurred during video capture stream? Robson, Adam, either one of you can jump in on that? ROBSON ADEM: Yes, so we do offer face blurring tool on Orbit. So this is something that we can provide. MAKAYLA KRUPP: Great. I have another question about the Wi-Fi setup that Spot needs to operate and stream data in real time. We're going to talk a little bit more about networking and infrastructure later. But either one of you, Robson, Adam, do you want to give a quick shout out to how that works? ADAM COMOLETTI: Yeah, I can give a good answer, but Robson, I think you probably have the specs off the top of your head. So why don't you go ahead? ROBSON ADEM: So we provide all the networking specs in our support center, so you can totally get all the details. But in short, Spot is designed to work with all kinds of networking modalities. That includes Wi-Fi, ethernet, LTE, radio networks as well. MAKAYLA KRUPP: OK. Orbit question, is Orbit on prem or cloud? ADAM COMOLETTI: I can take this one. So Orbit is both. So we have three options for Orbit. We have on-prem, version which is we sell you the server with the Orbit software on it. We have the cloud option which is through AWS. And then we recently released our VM option as well if you would like to host it yourself. MAKAYLA KRUPP: Great. We had a question come in pre-submitted about adding additional sensors to the robot. Adam, do you want to talk a little bit about the kit of sensors that we recommend and any options outside of that? ADAM COMOLETTI: Yeah. So the robot you see on the screen right now in the slides, that's our standard industrial inspection setup. So we have the Spot Cam on the front, our acoustic imaging payload in the center, and then our EAP 2 in the back. And that's our standard sensor package. We also like-- we talked about have the BLK ARC 3D scanner, like certified as a payload on Spot. Other than that, we do have an open API. So if you're good at programming, you're good at communication, you can integrate another payload with Spot. It won't be certified by us, but you can always do that. What we also offer, too, is we have our Boston Dynamics Consulting Group. So if there's ever a custom application you want to do a Spot, they can work on that for you as well. MAKAYLA KRUPP: Great. Maybe we'll just do a couple more before we keep going here. Robson, I think this one's a good one for you. Other than staff training, how do you ensure safety for workers when they're interacting with robots like Spot? ROBSON ADEM: Yeah, that's a great question. So Spot is a mobile robot. By virtue of that, it's going to be walking around your facility. So it's very important to us that safety of workers around Spot is ensured. We do have technological solutions to guarantee that. So we have a baked in obstacle avoidance that's going to make sure that Spot is not going to be running into any obstacles. In addition to that, earlier I mentioned our moving object autonomy. And we do have a special behaviors around people all. And Spot can stop for humans to clear the area before proceeding. And Spot also would be giving visual cues and audio cues about its presence, so those would be great resources to ensure that Spot is congruent with everybody working around it. ADAM COMOLETTI: Yeah. And the big thing I just want to touch on there is that the default behavior of Spot if a person or an object, say, a forklift is coming towards it is just to stop. That's the most predictable behavior. We don't want the robot trying to predict what you're going to do next. We want you predicting Spot. And Spot is going to stop and wait for you to either sit still or get out of the way. MAKAYLA KRUPP: Great. We'll have more time to answer some more questions at the end, but I'm going to go ahead and keep us moving here. The last one I think we can touch on segues just really well into what Robson's about to share. We had a few questions about reliability and Spot testing. So with that, Robson, I'll give it back to you. ROBSON ADEM: Awesome. That's a great question. So far, you may be wondering, OK, I'm already bought into Spot and Orbit. I like the story. Does that mean I have to test the reliability and usefulness in my facility? But in reality, that's already been proven for you. Today, we have over 1,500 Spots working around the world, collecting over 400,000 inspections a quarter with a collective mileage close to 80,000. We also have over 125 robots in our internal test fleet. And each robot in our internal fleet goes through several unique kinds of antagonistic testing to ensure robustness and reliability for both the hardware and the software. You may look at these metrics and only see how robust Spot is, but it's also delivering actionable inspection data on a reliable basis. The data that Spot is collecting is directly impacting customers and their business operation. Purina is a good example. They're using Spot to improve their processes for maintenance and beyond across their 20 sites. Let's look at how Purina is successful. We find that there are three things our most successful customers have in common. They define-- number 1, they define, measure and report on the value of the robot to justify the continued robot integrations. And number 2, they identify and engage site level stakeholders who will benefit from Spot's data early in the process before Spot is even on site. And the last one, which is also very critical, they have a way to make Spot's data accessible to those stakeholders within their existing workflows and systems. But first, there's a couple of things we need to get your site ready for the robot. The major two things are managing change. You're bringing the leading edge mobile autonomous robot to your site for the first time. It's very important to make sure everyone who will be working on Spot knows its purpose. This allows your team to know what to expect, including personal safety precautions. The idea is that the site shouldn't meet Spot for the first time when it's on the floor, doing its job. This is to ensure that we minimize any ambiguity about the purpose of the robot. We have lots of resources to help you with this. This is to increase awareness of Spot before it's on site, like safety posters, training videos, where some customers actually play this video in cafeterias where there will be large team gatherings. In addition to that, we have a dedicated support center to help you answer any questions you may have. Some customers even host meet and greets with Spot and also hold naming competitions which help with the morale and engagement of the site before Spot is on site. While you're doing that, we'll also be next to you working with the integration process. This includes laying the groundwork with the appropriate IT network infrastructure, as well as scoping your site to see where Spot can have the greatest impact immediately. Let's look at how this is possible. We do hundreds of deployments a year across the globe, so we have a process in place to help you get started, like most of our successful customers. In other words, we are with you at every step of the way. We have multiple teams at Boston Dynamics that are dedicated to helping customers like you go from 0 to 1, including project managers, field application engineers, customer success managers, support engineers that would help you along the way. Prior to Spot arriving on site, our team will work with you and the change management aspect we just mentioned. In addition to that, we'll get your rights-- you will get you site ready for Spot. This includes IT and site modifications. In addition to that, one of the pillars of successful robotic integration is defining impact and value before the robot is in operation. So we'll help you do that from day one. Typically, this prep phase takes about four weeks. After the robot arrives, our team will help you go from unboxing to operational value in just four weeks. This includes everything from getting operators trained to system installation and eventually, to site acceptance testing before we go operational. Finally, we do document the project results, establish operational support with our support and customer success teams. And this process essentially helping you go from 0 to 1 with your first robot operational. In addition, it sets the foundation to further scale within your sight and within your corporate network going up from 1 to n sites. Typically, customers who follow this process, in addition to the three pillars of successful robot integration that we've discussed earlier. See, ROI in two years or less, and this is captured in downtime prevention, time, labor savings, energy savings, and break down reduction. But let's break that down with a real customer example. Here's an example from a real food and beverage facility that we've been working for quite some time. You can see that they unlocked most of their value from breakdown reduction with over 180K US dollars. In addition, they do have energy savings and time savings that are each worth 30,000. This customer effectively defined and measured the value of Spot. Here's how they measure the value that Spot created. On the right side, you can see that they accounted for labor value per hour with Spot utilization rate to establish labor and time savings. To get to compute breakdown reduction, the customer looked at the number of capital equipment on site, the cost of replacing it, the cost of repairing it, together with failure rate of per year to get to percentage of failures avoided in order to get to compute break down reduction. Similarly, for downtime, we looked at cost of unexpected downtime per hour in hours of downtime avoided with Spot inspection. And finally, with energy saving, this is directly related to the compressed air leaks that Adam touched on, that we have a capability without leak detection. Let's look at another example. Here's an example from a real customer over an 18-month period. The difference here is that this customer has four robots, whereas the previous example had one robot. What is really interesting about this is that this customer used a similar single site ROI that I showed in previous slide to justify for additional three robots and in three additional sites. At the end of the 18-month mark, the customer was eager to measure the value of Spot across their four sites to justify additional purchase across their global sites. We help this customer in generating this automated report. This is done through our integration services and dedicated field application engineers and customer success managers. We touch base with customer on a weekly basis in a data-focused, value-first approach to extract the best impact out of Spot. In other words, we will actually help you generate these reports to track the impact of your robot. So you have the information at your fingerprint to share with your internal stakeholders and track progress of your robotic programs. Now that we have seen defining and measuring value looks like, let's look at a second pillar that contribute to successful robot integration. Looking at the value information, you may think that Spot is really just a maintenance tool serving your reliability engineering and maintenance teams. The truth is that there are so many more people within your organization that can benefit from Spot. For example, your line owner will see a reduced unplanned downtime. Your plant manager will be thrilled to see increased productivity and workforce upskilling. Your safety team will use Spot to keep people away from hazards. Your data science team can also use Spot to collect internal data to build its own internal AI efforts. This goes up all the way to your executive teams realizing their business goals. In other words, the value Spot delivers mean something to everyone. As such, it's very important to engage a various teams and personas early to define the value Spot brings to each team. Think of each of these stakeholders as potential advocates for your robot program. Let's look at how we can harness the data. With the proper site level buying across many teams and personas, you have now-- You are now on your way to establish a network of several end users of data of the Spot that is collecting. This network is maximized in the ways in which you can harness the data. Spot can tell you maintenance tech what 20 repairs they need to prioritize this week, but it can also tell your safety officer that there are six final issues that were resolved before the auditor showed up. Traditionally, all this data is siloed in the industry. But with Spot and Orbit, our data's platform, so you can get to you can get it to the people who need it the most. Orbit is at the center of that platform. That's where all the data Spot collects is funneling. And you can certainly work with Orbit to analyze and trend data and triage anomalies. But at the same time, you can also use other mechanisms that we have through our API and webhooks to make sure you can pipe the data straight to the places where your team is already working. As I mentioned before, one of the three common contributors to success with Spot in Orbit is customers having a way to make Spots data accessible to stakeholders within the existing workflows and systems, such as facility, management systems, [? and, yes, ?] EM things of that nature. Now let's hear from Adam about customer case studies that clearly showcase that we have talked about so far. ADAM COMOLETTI: Yeah. Thanks, Robson. So we've learned about Spot. We've learned about how to make Spot successful. Now let's take a look at where that often begins. And that's building the business case for Spot. So this is one of the core ways my team, the sales engineering team, supports our customers early in the process. And that's helping identify where Spot can deliver the most value and quantifying that value in a way that aligns with your operational goals. So for the case study I'm going to walk through, we'll be looking at a beverage manufacturer. They produce over 1.2 billion cans annually. And like many facilities in the food and beverage space, they're focused on keeping uptime high and risk low. The areas they identified for Spot are the ones we commonly see across this industry, and that's the packaging area, utilities, and production. And these are environments with a lot of moving parts literally and figuratively. But an overview of the facility. So as part of our evaluation with the customer, we identified approximately 2,700 inspection points in total that Spot could be assigned to monitor on a weekly basis. And these inspection opportunities span across the three main areas we discussed. So of those 2,700 inspections, we identified around 320 visual inspections, mostly AIVI, including things like analog gauges, indicator lights and fluid levels. There were just over 1,000 thermal checks covering assets like motors, electrical cabinets, and pumps, where temperature changes can indicate early signs of failure. And then there are roughly 1,200 potential leak detection points. And when we say leak detection points, we mostly mean where the fittings are because that's where leaks are most common to occur. And finally, we identified 220 mechanical inspection points. And this is where Spot could listen for the abnormal vibration or bearing noise. So this gives us a clear picture of where Spot could be deployed and the volume of inspection tasks that could handle, which becomes the foundation for building a flexible, scalable business case. So with those inspection opportunities identified, we worked with the customer to map out a few different integration strategies, each with different scopes, associated inspection volumes, and potential ROI. So as you can see, with option 1, that's a standard deployment covering three packaging lines, about 300 inspections per week. And this provides a solid proof of value. But of course, 300 inspections is not enough. This is going to result in a longer payback period of just over 38 months if we just stuck to those 300 inspections. So option 2 expands that coverage to all 10 packaging lines, reaching 1,200 inspections per week. And this brings the ROI down to 8.6 months, thanks to the higher impact across labor and downtime reduction. Now, option 3 is biting off everything at once, let's deploy the entire facility-- 10 lines, utilities, processing, taking on all 2,700 identified inspections. And this approach drives the greatest annual value nearly $1 million, and yields a payback in less than five months. Now, some key takeaways from this. As Robson mentioned before, most customers we speak with aim for a payback period of 24 months or less. As you can see, even the mid-tier option for this customer far exceeds that goal. Now, second, the value Spot delivers really scales with the inspection volume, as you can see, because you're spreading similar fixed costs across more inspections. so it's increasing the probability of finding issues which can tie back to the ROI. And unlike fixed sensors or manual rounds, Spots flexibility lets you unlock ROI from multiple categories. So just like a vibration sensor only does vibration. Well, Spot can do air leak reduction, labor augmentation, downtime avoidance, mechanical reliability, all with one platform. So really, whether you're starting small or going site wide from the start, we can help you build a strong business case for Spot. And really my team is here to help you build it. So just to wrap things up, I want to take a moment to show you where Spot fits in within the broader Boston Dynamics ecosystem. So we're building a family of robotics platforms that bring intelligence and autonomy to the physical world across industries, environments, and workflows. So on the left, of course, we have Spot, which you've seen in depth today. And Spot is our autonomous mobile platform for data collection and industrial inspection. Spot is deployed across thousands of facilities worldwide and available today. Now in the middle, the one everyone gets the most excited about is Atlas, which is our next-generation autonomous mobile manipulator. And Atlas is being developed to handle complex human-like manipulation tasks, combining mobility, dexterity, and perception. It's not commercially available yet, but it represents the future of mobile manipulation in unstructured environments. And last on the right is Stretch, our flexible warehouse automation solution. So this solution is purpose built for trailer and container unloading, with new applications to come soon. Stretch is also commercially available today and already making an impact in a high throughput logistics environments. And together, these platforms powered by our Orbit software will represent a connected intelligent automation ecosystem, one that bridges the gap between the physical and digital worlds and helps businesses unlock a smarter, safer, and more scalable future. But all right, with that, I just want to thank everyone for spending some time with us today. I hope this gave you all a good sense of what's possible with Spot and where we're headed. But I'll turn it back over to Makayla now to take us into the final Q&A. MAKAYLA KRUPP: Yeah. We've got lots of questions coming in the chat, so keep them coming. And if we don't get to your question right now, we'll have folks following up after the webinar. So keep using that Q&A. We had one question regarding data collection. Is it possible to leverage machine learning models to train the robots for task-specific inspections? For example, validating whether a service was executed according to a predefined plan, in cases where deviations or quality issues are detected, how are these insights captured and integrated into reporting or analytics platforms? Robson Adam, either one of you want to take that one? ROBSON ADEM: Yeah. I'll start an item can add some flavor to it. So Adam already touched on AIVI, which is breaking grounds for us, delving into harnessing the data using AI. So we will be excited to hear more about your use case for that application, as we're continuing to make our models more flexible to other applications. ADAM COMOLETTI: Yeah. And what I'd add to that is I envision a world where if the AI inspection detects something, we have a behavior set up. So hey, you're alerted of this, so do this behavior. So all that would be is some back end changes in the software. And I don't see why that wouldn't be possible with Spot. MAKAYLA KRUPP: Can Spot detect streams of fluid? I work at a place that has a lot of leaks. ADAM COMOLETTI: Depends. If it's water, probably not. That's pretty transparent. If you have oil leaking, much better chance of that. MAKAYLA KRUPP: We get this question a lot. What is the expected life expectancy of the robot on average? ADAM COMOLETTI: Yeah. So I can touch on that. So I wish I had a good answer. The real answer is that, like Robson touched on, we have our internal test fleet. I think there's only been one robot from that internal test fleet that's been end of life. Like, can't run anymore. So that's been 4 and 1/2, five years. So really, what I can say is at least five years, because, again, none of those, except for the one that we beat the crap out of, actually hit end of life. So we can tell you at least 5' years. In most accounting for depreciation, uses 5 to 7 years. MAKAYLA KRUPP: Perfect. How does fleet management work if you have more than one robot? And maybe we can hit on the difference between one site, multiple sites. Robson, do you want to talk a little bit about Orbit fleet management capabilities? ROBSON ADEM: Yes, absolutely. So Orbit can host multiple robots across your sites. In addition, we also have enterprise Orbit, which we released in 5.0, which is the latest software we release for Spot on Orbit. Through Enterprise Orbit, you can also network multiple orbits to get to the information you need across your sites. So we have the ability to manage fleet in different levels, at site level and also at account level and corporate level. MAKAYLA KRUPP: Perfect. We had a question pre-submitted about how Spot operates in rain. Adam, do you want to hit on that one? ADAM COMOLETTI: Yeah. Rain usually not a problem. Spot is IP54 which means it's water and dust resistant. Now, again, that's not waterproof. So you can't throw Spot in a pond. But rain is generally OK. Heavy rain, I'm a little tentative with that. We have increased or improved our hip ceiling. On top of that, we also have a weather scheduler integrated into Orbit. So say, before Spot goes out in the mission, it's going to check with that weather scheduler. What's the weather? Is this within my parameters? Yes or no, and that decides if Spot goes out in the mission. On top of that, usually for our more remote customers, think utility companies that don't have people on site, but they're outdoors, they'll purchase our Spot House, which is an earthquake-proof enclosure that has Spots docking station. It has air conditioning. It has a door that opens with RFID. And that's how we combat the outdoors right now with Spot. MAKAYLA KRUPP: Perfect. We had one question about how Spot is controlled. So Robson, maybe do you want to hit on the different options that you have for operating the robot? ROBSON ADEM: So Spot has different modalities and clients that you can interface with. The first one is the tablet. The tablet, the way we like to see it is a line inside of control. The tablet is used to drive Spot around for teleoperation within line of sight, but in addition, we also use the tablet to teach Spot on autonomous inspection. Similarly on Orbit, Orbit has teleoperation, but an addition is a larger software that also includes fleet management, data analytics, and also piping data to other parts of your facility systems. So those are the two major ways to connect Spot. MAKAYLA KRUPP: Yeah. This question was specifically giving an example of being far away from the robot. So in that case, Orbit would be your solution. ROBSON ADEM: Yup. And we have long-range radio tools that also we use in other markets as well. MAKAYLA KRUPP: Yeah. Great. Another one in the chat here. Why was Spot given a dog walking design? Why not wheels or tank-like wheels? What's up with the dog format? ADAM COMOLETTI: Yeah. I mean, so legs, the world's built for legs. Every facility we go into stairs, everything like that. Tracks are good. If you're just staying on one floor, yeah, tracks might actually be a better option. But a lot of these facilities, a lot of the places Spot goes into need the advanced mobility that you get from legs. Now being named Spot and looking like a dog, you can imagine that just helps with adoption. You see Spot, you're like, oh, he's so cute, she's so cute. That helps. But it's really also just a form factor that's fit well into the world. MAKAYLA KRUPP: Yeah. And like Robson was talking about earlier, the legs enable Spot to do stairs, navigate around terrain, objects, et cetera. Robson, you hit on this in the webinar presentation already. But maybe just to reiterate, what's the typical deployment timeline for Spot and Orbit in a traditional facility? ROBSON ADEM: So the way I would like to answer that question is in two phases. The site prep phase, basically getting you your site ready for Spot. Then we typically budget about four weeks. And this gives the sites in general enough time to put together the IT infrastructure around Spot and also doing site modifications, including door automation. And once Spot arrives, we go from unboxing the robot to value within four weeks. So in total, it would probably take you 6 to 8 weeks to get Spot up and running, start from to end. MAKAYLA KRUPP: Perfect. We just had one come in about door opening and elevators. How do Spot deal with doors and elevators? ADAM COMOLETTI: Do you want to take that, Rob? Do you want me-- ROBSON ADEM: Yeah, I can take. I can take the automation. So we actually in 5.0, for our current software we released a couple of months ago, we're allowing customers to interface with door automation with Spot. Spot, if their door is automated, would check if the doors open and proceed. If it's not open, it would also let the administration know that Spot is stuck and it would reroute back to its dock. So that behavior has been a fantastic with automation. With elevator, there are customers who've experimented with this. But at this moment, this is out of spec for us in general in our industrial facilities. MAKAYLA KRUPP: Another one came in here. After a successful site deployment, are there greater conversations with customers to explore similar sites for expansion, or is that something that the customer has to initiate with you? I know we covered some of this in the content, but is there anything else you'd want to add on that, either of you? ADAM COMOLETTI: Yeah, absolutely. So towards the end of your integration, you'll be looped in with customer success. Now your sales team, the sales engineer and the sales manager, will be along during the integration, will be more in the background letting Robson's team do their thing. But as soon as you start getting to the end of that integration, yes, we'll re-engage. We'll have conversations with you and see, hey, how do we scale this solution now? MAKAYLA KRUPP: Great. ROBSON ADEM: And we do that by providing you the data I showed earlier with generating the impact story with numbers. MAKAYLA KRUPP: Perfect. I think we've got time for just a couple more here, so I'll throw two more out there. What are some of the limitations of Spot and Orbit today in terms of data types and plant conditions? ROBSON ADEM: Yeah, I think this is very environmental dependent. Spot goes around lots of facilities. We've seen some limitations in dark environments when it comes to data collection, and this is with regards to visual inspection. And there are some targets in the industry that are just simply not accessible with Spot, either tight stairs or tighter catwalks, or just simply a gauge that is 30 feet up high. MAKAYLA KRUPP: And I think to wrap us up, Adam, I'll leave you with the final question here. You hit on or hinted on this with your last slide, but do you see Spot evolving towards performing basic autonomous repairs or interventions in the future? ADAM COMOLETTI: Yeah, don't I wish. But where you see with Spot, Spot with an arm, it's just one arm. It's very hard to manipulate things, put an arm behind your back and try to manipulate things. Turning a valve, you can do that. But complex manipulation, no, it can't do that. So Spots are always going to be our mobile data collection tool. Atlas is going to be our mobile manipulator. So when you say, oh, can it do repairs? That'll probably be Atlas. So envision a world where 5 to 10 years from now, Spot finds a problem and then sends it to an Atlas to go fix the problem. MAKAYLA KRUPP: Perfect. Yeah. This is the future. We do have use cases for Spot manipulation and other industries. So if you're in the public safety space and have some less intense manipulation tasks, that's a great use case for the arm. But Atlas is definitely where we're headed for manipulation and industrial settings. ADAM COMOLETTI: Yeah. What I would say to that is Spot is great at if you're controlling it for manual operation, there's some autonomy there. You can just click on a doorknob on the screen and it will open the door. There's some autonomy there. But the autonomous manipulation set it and forget it, go do this manipulation task, Spot's not really good at that. MAKAYLA KRUPP: OK. Great. Actually, I'm going to leave one more question to wrap us up. We had a couple questions about comparing Spot to manual inspections, fixed IoT, building that business case. We talked about it at varying points in the webinar. But either of you, do you have any final thoughts on how you justify Spot in regards to those other options? ADAM COMOLETTI: Yeah. So I mean, the numbers have to make sense. Like I said, we go through that business case exercise, the numbers have to make sense. If you have three or four technicians going around doing rounds and they could be off repairing items, that's more valuable. Spot can augment that labor task and go let those technicians do those more high value tasks. Now fixed sensors, a lot of people say we're competition with fixed sensors and fixed sensors are our competition. I don't see it like that. I see a world where we coexist. Because fixed sensors, they do give you that 24/7 data. Every second, they're pulling that data and collecting it. And on your top 10% of assets, you should have fixed sensors on those. But the other 90% that only need to be checked once, twice, three times a week, just have Spot inspect those. So now you're not spending all that money outfitting your whole site with sensors. You're just getting those on your most critical pieces of assets and then Spot inspects the rest. MAKAYLA KRUPP: Beautiful. I think that's a great place to wrap things up for today. So if we didn't get to your question, I'm so sorry we ran out of time. We will be following up on those. So if you've got them lingering in the chat there, someone will be in touch with you. And you can also use these links to find more information about Spot on our website. So thank you, again, everyone, for joining. This has been recorded, that will be sent out to you shortly. And look for our next one in the coming months. Thanks, everyone.