Leveraging Robotics for Smarter Perimeter Security Inspections

- To kick things off with around of introductions, let's start with Paul. - Hello, everyone. I'm Paul Cisternelli, and I'm part of the Enterprise Sales Team at Boston Dynamics. I've been here almost four years working on selling the Spot robot into the market and have been selling it to a number of different markets. And in the past year, I've been focusing on really kind of elevating our presence in the greater security market, so this is a great opportunity for me to introduce the product to the market. - Hey, everyone. My name is Damon Henry. I'm a CEO and co-founder of Asylon Robotics. Started Asylon about nine years ago, and we've been chasing the ability to operationalize robots in the physical security space. - Hi, everyone. My name is Matt Pascal. I'm the corporate security lead for Bayer Crop Sciences West Region, which covers Hawaii, Arizona, and parts of California. - Awesome. So yeah, let me get started. I thought before we jump into the security solution using Spot and everything, I'd talk a little bit about Boston Dynamics and our robotics in general. A lot of people have different perceptions of what robots are, and why people use them, and why they're getting more and more publicity over the past few years. And all of that's true. Robots are very cool, they're very innovative, they're very exciting technologies to deal with. But what we found in the marketplace, and we've sold probably nearly 2,000 Spot robots over the past four years, what we really understand is that the reason why people are purchasing these robots is for a very practical use, and they're trying to solve a problem that's nearly universal around the globe today, and that's the problem with the labor deficit. There is a lot of shortages in the labor force out there that are causing pains with our industries and adding to the cost of goods sold. And there's many reasons for this, whether that's an aging population, or people changing their work preferences, people not wanting to do certain types of jobs. There's migration, and climate change, and lots of things going on. But regardless of that, what's happening is you see that the labor supply is going down and the demand for these types of functions that that labor would provide is going up. And when you have supply and demand crossing those lines, you're going to have an increase in costs, and that's what we're facing in the world today. And some of the industries where really see that is these labor intensive industries, such as security, which we're going to talk about today, manufacturing, where people need inspections done on a routine basis day in and day out. Those jobs are not getting done because there's not labor to do it, and that's causing problems down the supply chain. Logistics, and warehousing, and things like that. So there are a number of industries where we're finding our robots are being utilized for very practical reasons, and today we're going to talk a little bit about the security element of that. So on the next slide, let me chat a little bit about the size of the market that we're talking about here, and a lot of you folks probably already know these things. But the market for security guard services and I'm just talking about guard services at this point, is around $50 billion last year, and it's growing at nearly a 6% compound annual growth rate. And that will continue over the next few years. And that tells you that demand is going up. People want to avoid any disruptions in their operations, they want to avoid leakage, they want to avoid any potential downtime to their operations, so they're providing more funding to security to avoid those problems. In the US alone, there's about 1.5 million security guards employed in the US and those jobs are very hard to fill. So we're having a challenge in that the number's going up on a year-to-year basis. Some of the dynamics that I've heard from the marketplace and I've read online, is a turnover nearly 100%. So that tells you that for every guard that's hired, within the year, you're going to replace that guard. 40% of the guards that you hire are going to probably leave and find a new job within four months. If you consider that the time to recruit, to onboard, to train, and then finally deploy an individual, that individual likely to just leave the job. So now you're just starting over and over again. That's all driving the cost. So again, the demand is going up, supply is going down. We are headed for an inflection point that you're going to have to create a dynamic shift in the underlying business models because the current model is not going to sustain. Either you're just going to have to keep charging higher and higher wages for these employees or you're going to have to find alternatives to provide the same level of service, if not, improve service at a lower cost. And that's really on the next slide, what we're going to talk about, how robotics is affecting the security market. So I'll just talk about three key areas that we're going to talk about today, and then I'm going to hand it off to our friends at Asylon robotics to show how this is actually working in practice. But the three key areas I wanna focus on is cost efficiency, because this has to be cost efficient. That's the whole reason why behind this. If this is a more expensive option, it's not a good option. At the same time, effectiveness, enhanced effectiveness. So even if we are saving a few dollars on the hourly wage, if it's not as effective, then it's not serving the purpose. So we're going to talk about those two things with robotics. And then finally, and perhaps less RIO-oriented, but just as critical, is the safety element of this. So let's dive into each one of these just a little bit. In terms of cost effectiveness. As we all labor is expensive. I just told you we have a supply/demand problem. From our economics courses 101 in college, we all know that when supply and demand cross prices go up, we have problems. So we need to better utilize our high-value resources. So not only take the labor problems, but change those people that we're using to do other functions and on their own, whether that's assessing deployment reaction, things like that. So robots can handle these routine duties. They can do it over and over again. They don't argue about it, they don't take coffee breaks, they don't take lunch breaks. They just continue to work as demanded. So this frees up those resources to do other critical work. So reliability becomes a big issue. So in addition to the labor itself, where you can actually drive the labor hourly costs down by using robots, maybe somewhere between 20% and 25% of the hourly wage of a physical labor to a robot, you also have additional benefits, like liability reduction. You don't have to have liability insurance. Over time reduction. The robots can do it over and over again. You don't have to pay them overtime. Increased uptime, which gives you a higher yield for your operations. All of these things develop into the cost efficiency of driving robots. The second big one is this enhanced effectiveness, and that really gets to the concept of the robot itself, where you have a four-legged robot that is enabled to traverse multiple types of terrain. In this example in the video you're watching, you'd see the robot climbing a very steep incline with loose gravel rocks, and the robot is actually figuring out where to place its legs. And as the rocks give way underneath its feet, it automatically rebalances itself. So the robot has the capability of getting anywhere it needs to on a perimeter surveillance patrol. So if that ground is level or not level, there's curbs involved, there's dirt tracks, there's grassy tracks, the robot can traverse that very easily. It also has an open API, which allows you to put different types of payloads on back of the robot. So those might be cameras with night vision and optical zoom lenses that allow you to do certain types of surveillance as the robot's walking around. And then as the robot's walking around-- excuse me-- all of that information is streamed live back to a command center so that your higher value employees can take a look at this and do determination of what's going on. The other thing about robotics is all that data is electronically documented. So it is time stamped. It tells you exactly what it saw and when it saw it, and then you can integrate that with your other core systems. At the end of each patrol, the robot will find its way, make its way back to its DogHouse, which we'll show you in a minute, and then recharge, and as it's recharging, you can upload core data that it captured along its path. When it's done recharging, it'll get up and do it all over again. And this is really the benefit of having a cost-efficient, but even a more effective robot in terms of delivery. And then the final thing I want to chat about is a little bit-- which is near and dear to everybody's heart is safety. So what the scenario here is, you might have a situation where an alarm has gone off from a door being opened or a fixed camera might have spotted something that your security team wasn't there for, and your first response must be, hey, let's send out a guard to check this out. So instead of sending out a guard, a human, into this without knowledge of the situation, we can send the robot as the first responder without risking personnel. That in itself has the liability reduction that allows you to send that robot out, get eyes on the scene so that the command center can accurately determine the proper level of response. And that's important because the proper level of response might be, hey, we need to call the police right now, or, we need to dispatch an armed guard, or, you know what? It was just an animal passing by, we don't have to do anything. So instead of dispatching somebody when it's not needed, we can make that determination. And then final thing in safety, what the robots provide is what we call a passive deterrent. So a lot of times when I started up, a lot of people have different perception of what robots are, and in this case, it kind of works to our benefit. So as a robot approaches a perpetrator, the first thing they're going to do is stop and they're going to get distracted because they don't really have a full understanding of what this robot is capable of. What's it doing? Is it tracking me? Is it feeding information? Is it going to do something? At worst, it's just stopping you and delaying the actions of the perpetrator. At best, it can actually stop the perpetrator who will then back out and leave the situation. So these types of things all drive safety. So in my next slide, I just summarize this. We have a very large market opportunity here, and the market is facing a inflection point where the underlying business models are going to have to be adopted and robotics is the best way of doing that because of cost efficiencies, effectiveness, enhancements, quality, and safety. Those three elements are key. So with that, let me hand this off to Damon Henry of Asylon Robotics, who will dive into how he's using our robots to do this. - Awesome. Thanks, Paul. Appreciate it. And great overview. I think definitely a testament to both the Boston Dynamics and the Asylon teams and their dedication to this use case. And so again, my name is Damon Henry. I'm a CEO and co-founder of Asylon. Our vision as a company was to advance technology to create a safer and more secure world. The way we do this is by providing physical security as a service leveraging robots and AI. Today, we offer two different products. The first, which you just saw, a drone dog, is our ground asset, and then we also have a fully-automated Drone-In-A-Box platform. And the way we do this is, we deploy our robots to the facility. These robots do the three D's, the dull, the dirty, the dangerous. So they take over a lot of the fence line patrols, the mundane security tasks, but then at the next level, we integrate with the existing physical security layer. So if you have guards, cameras, access control, alarm panels, we can actually integrate the robots into those devices so that they can become that first responder and get eyes on the scene. Do that double verification before you expand the manpower, which is so critical to us from a time perspective. So now you have this integrated solution. You have these mobile cameras moving around your sites, and we push all of that up to the cloud through our software, DroneIQ, and then ultimately back down to our customer. From this, they're able to see the real-time security video feeds, the analytics. And what's really come out of this that's interesting is you actually get a reporting tool, a metric tool, as to what's going on around your facility. With the robots being these active mobile devices on your site and then integrated with all of these other security devices, you get this holistic view of what's going on around your facility. So we're able to essentially draw heat maps, do that, show the trends over time, and do that analysis so that you can start to morph your security posture to essentially de-risk your facility and make it safer from an operational standpoint. So in the next slide, kind of diving into the DroneDog platform itself. So again, our goal is to operationalize these robots into the physical security space. And so with DroneDog, what you have here, you have the base Spot platform for Boston Dynamics. On top of that Asylon, we've developed the PupPack. This is our Security Payload. It has both daytime and nighttime cameras with zoom and tracking capabilities. It has onboard edge computing so that we can run our classifiers and our security analytics. And then ultimately it has everything needed to turn Spot into this automated, remotely controllable security asset on the ground level. In addition to that, we've also built the DogHouse which you can see here, which houses the Boston Dynamics charging unit. And so what this really allows it to do is Spot deliver a security device in the variety of environments that security operates in. It's a 24/7 business. It doesn't really matter what the season are or what the time is, the assets need to operate. And so the DogHouse enables the Spot to return and recharge regardless of the environment. So whether it's 100 degrees in the summer or snowing in New England in the winter time, we're able to continue those security patrols and provide that 24/7 coverage with this device. And so looking back at Asylon as a whole and this will be a common theme, particularly as we get into the discussion with Matt. We don't believe that robots are the end all be all in security. We believe that it takes the hardware and the software layer from the robots, but you have to merge that with the human expertise component. That's really how we're going to redefine security and get to true enhanced security. Asylon, as a whole, our commercial business, we've done over 150,000 robotic security missions at this point. We're happy to provide services for a variety of Fortune 500 clients like Bayer and others, really protecting high-value assets, critical infrastructure, and other assets within our supply chain. On the defense side, we have active contracts with the US Air Force and Space Force, who have a really parallel use case to the commercial world, which is we have lots of lands, lots of fence line to protect, and it's really hard to do with traditional measures of just people and fixed CCTV cameras. So us being able to bring robots into the space coupled with that human expertise element, is enabling us to enhance the security, be more effective, be more consistent, all while saving costs to the overall security posture. So finally, some interesting stats here on the DroneDog platform. We've done over 100,000 missions with our ground assets. This turns into about 90,000 miles of fence line patrolled with these assets. I just want to pause here and talk through how big of a milestone walking 90,000 miles is with the robot. I think it's a testament to the Boston Dynamics team with their platform, and Asylon's team on the operational side, to do this level of patrolling and consistency, operating day in and day out, hour by hour, to ensure that we're providing the 24/7 capabilities and the service levels that are required in physical security, which just never shuts down, right. Some other interesting things here. Our robots have been involved in over 1,000 security or safety events at this point, everything from active fires. We've had some fire prevention to train robberies even, and some active thefts. And so it's great to see the expansion of robotics into this use case becoming more normalized. For us at this point, almost on a daily basis, the robots are recognizing and capturing security and safety events for our customers, which is incredible to see. And then I talked about the human elements. On the back end for us here in Philly at our HQ, we have what we call the RSOC, our Robotics Security Operations Center. We have 30 plus trained analysts currently on staff on all shifts, and these are the folks that are logged into the back end of the robots. They're monitoring the security video, they're ensuring that the robots are operating at the tempo they're required to operate at. If a robot gets stuck, anything like that, they're able to remotely log in, get it running again. And so this is that other critical piece, right? Doing the human verification. So that when we contact our customers like Bayer, you have that human element. You can react very quickly, and effectively, and efficiently, to what's going on, particularly if it's an emerging situation at the customer site. So with that, I'll turn it over to Matt at Bayer to talk through their journey deploying robots for security. Thanks, Damon. Hello, everybody. I'm calling from Maui today. I'm going to reintroduce myself again. I'm Matt Paschoal, I'm the corporate security lead for Bayer Crop Science's Pacific West Region. And I'm going to be discussing really how we've been utilizing Boston Dynamics and Asylon Robotic's DroneDogs at our Hawaii sites as a component of our layered physical protection system, specifically in the area of perimeter detection. I think to really illustrate that, I kind of need to paint a picture of who we are and what we do. So you know, as Bayer's proprietary security organization, our vision is pretty straightforward, right? Our vision is to support the continued success of our business operations and long-term growth, really by providing the best-in-class strategic security services we can for the company. And how do we accomplish that? Really, to make that vision a reality is we have one primary mission. It's to protect our employees, our assets, our real estate property, our intellectual property, and even our license to operate from all threats, internal and external. A lot of times people ask, why do you need security at a farm, right? And I mentioned, we're a crop science. So what we do here in Hawaii is we grow corn. That's our primary product. So why do you need security at a farm? And it's a really simple answer. Hawaii plays a critical role in Bayer's corn products pipeline, which is an integral part of the US food supply chain. So what makes Hawaii so critical is really our weather out here. The great weather that we have that everyone is known to come and love when they come over here on vacation, it creates a multi-season production environment for our corn products. We can really get anywhere from three to four growing seasons a year. Now, you compare that to the mainland where you get maybe one viable growing season a year, and you can see the huge competitive edge that gives Bayer, especially for introducing new hybrids to the market. Just for scale, I'll let you know that 100% of all domestically-sold corn hybrid seeds start in Maui soil, and 90% of all internationally-sold corn seed hybrid corn seed started in Maui soil. So just given those stats, you can see what an important role our farms here play, and how valuable the crops that we produce are, and so we absolutely need to protect them in any and all ways possible. So that's a little bit about us. And now I'll talk about how do we come to a point where we wanted to explore or start looking at robots for our security solutions. Really, I think like most corporate security organizations, it comes down to finance, right? Just like a lot of companies, we face significant resource constraints. Limited budgets, increased competition for available funding from other divisions, they really can hinder our ability to implement comprehensive security measures and respond to threats effectively. So couple that with increased geographic areas of responsibility here in Hawaii. We purchased a couple new farms and we did not get a correlated increase in headcount to accommodate for those increase in acquired land, so it creates an environment of having to do more with less, right? That became our mantra for corporate security, is doing more with less. And to be honest, what Paul spoke to earlier, even if we were to get a headcount approved, it's very difficult to find people fill those roles. Just having people available, the high rate of turnover, it's just really tough. So that situation, it kind of lended itself perfectly to what I was seeing in the robotics industry, especially in the realm of security, and how it could be a viable solution for us. Being able to do more with less really requires that a security leader stays ahead of the-- or organization, stays ahead of the curve, kind of keeping abreast of rapid advancements in security technology, as well as the emerging threat landscape. And having this mindset will help us identify opportunities for us to leverage tech. That can help us gain operational efficiencies in a lot of areas, but primarily what I was interested in was headcount. So giving us those operational efficiencies and still allowing us to successfully fulfill our core mission. And so in this vein, when you're talking about technology, I happen to-- I happened to look at security robots and thought just from all the benefits that were just stated by the previous two presenters, they would be a fantastic option for us to explore. So either if they were going to be automated, or autonomous, or piloted robots as part of our physical protection system, but was interested in using robots as a force multiplier for our team and really help provide a more enhanced perimeter detection system. We trialed, and I'll be honest with you, Asylon wasn't our first go-to. We tried a couple of sources, we tried a couple of endeavors. We call them proof-of-concept trials for a couple of different wheeled-platform robots that kind of really promoted an autonomous or a fully autonomous solution. But to be honest with you, they never really provided the reliability and dependability that we are looking at that we needed, we required in order to make them a real viable option. So I did have Asylon in my mind, I saw their product, I thought it was a really interesting, so we decided to give them a shot. But those first two failures kind of-- I wouldn't call them failures, they're more learning experiences. They really helped me and our team kind recognize that there is no perfect autonomous solution out there. You need that-- for us, you need that trifecta, the robot trifecta, which I think Paul discussed earlier. It's having the hardware, the software, and dedicated people to run this to really make a successful robotics program, and Asylon really just checked the boxes in all of these areas. So we decided to give their hybrid system a shot. Our robots are piloted by highly-experienced third-party SOC operators. And it still gave us the head headcount efficiency we were looking for. The net result was we weren't operating it, we didn't have to drive them, so we decided to give it a shot. We got a trial set up here in Maui back in January of this year, and a just absolutely phenomenal company. I can't stress enough how professional these engineers [INAUDIBLE] and everyone in that company was. They got us-- they got the DroneDog, a DroneHouse out here to Kihei, and they helped us. Sat down with us and helped us create a custom patrol-- well, perimeter patrol maps, really, that are going to give us that enhanced coverage that we're looking for. And they took into account our views, what we deemed important, what assets we needed to keep my eyes on, what areas were at higher risk based on our own security vulnerability assessments. So they were able to customize patrol maps for us that the dogs would follow on a daily basis. And as a result, we were really able to achieve a higher saturation rate for patrols. With the dogs continuously walking around, they're really saturating our campus way more than we had with a guard in a truck taking a pass by every once in a while. So that result was more tours, and the drone dogs fit very nicely into our overall-- You look at our overall security taxonomy, we're talking about deterrence, detection, delay and respond. So these DroneDogs, as Paul mentioned, people don't know how to act. They don't know how to react to robots. Maybe movies do help us. Maybe they think it's a Terminator. I don't know. But when they see the lights light up, the strobes going on, and they're walking around. And I love these robots, but they do-- they can look a little creepy in the beginning. So they kind of provided that-- still provided the deterrent factor right for us. And absolutely with the security payload that David and his team put on these robots, it was just incredible. The cameras are fantastic, with visible light spectrum infrared thermal, we're able to see in complete darkness. So they really gave us an enhanced capacity to see things occurring around our sites. And we were able to pipe those feeds, those camera feeds. We'd pipe them right into our SOCs so our RSOC operators can actually see everything that's happening. They can everything the dog's seeing, either in real time or even from past-recorded missions. So again, it allows us to bring our human talent inwards, gather data from the field through excellent solutions, such as the drone dog, and helps our operators do their triage right and initiate the proper response to this natural threat-- so a false alarm or whatever. We're able to bring that discerning human factor into the SOC in that town in board. I mentioned head count efficiencies multiple times. So I'm going-- I want to put-- take a minute to put some really hard figures behind that statement. When we first indicated Asylon, we wanted to bring them on board as a possible solution. We had to do a financial justification document just like we do with every project that we bring to the table. So based on our-- looking at it from a head count perspective and based on our current wages, our payroll liabilities-- benefits such as bonuses and incentives, and then even incorporating fuel and maintenance costs into the equation-- we save-- this is a real cost-- we save $90,000 per year per dog that we deploy-- $90,000 per year per dog we deploy. That's a huge figure. That's a figure that the business could get behind. And so we were able to greenlight this trial and get it going. And this is operating at the 112-hour per week model-- so basically, 16 hours a day, seven days a week. Another benefit-- huge benefit was-- we saw as an upside-- was the reduced potential for liability. These robots create a much safer environment. I know Paul mentioned this. But I'm talking a real-world scenario here around our farms or in our fields. We have a lot of unstable land, gravel, loose dirt. It helps being able to deploy these dogs in situations or areas like that. It can really help reduce injury to our security guards that might otherwise have to patrol those areas on foot, not to mention, if we get surprised by an intruder or whatnot. So it creates a safer environment for our guards here. And then lastly, we live in Hawaii, and it's a beautiful environment. Everybody wants to live here. Everybody wants to come here. It's a great place for business. But we have a huge responsibility to take care of the land-- or we call it the "aina." I'm native-- born and raised here in Hawaii-- and I love this place. And we take it very seriously being good stewards of the land and our resources and taking care of the environment. So these drone dogs, being that they're electric-- and we're not using fossil fuels and idling patrol truck 112 hours a week-- we actually did the math. And we calculated that we saved 70%-- we actually got a 70% reduction in our CO2 emissions annually when compared to idling passenger trucks using fossil fuels. That's huge. It's huge. And it's something that our company could grab a hold of and really promote in our sustainability efforts. So anyway, that's who we are, what we do, and what led us to using-- to evaluating robots as a potential solution for our facilities. And I guess with that, I'm going to hand this back to Damon so he could discuss his slide. There you go. - Yeah, absolutely. Thanks, Matt. Appreciate it. - Sure. - Yeah. So physical security, it's a hands-on business. So we really wanted to take a few minutes to talk through, how do you actually set this up? How do you integrate a robot into your physical security program? And so Matt mentioned, you have the hardware, the software, and the human element. And so from an architecture standpoint, you've got the robots. That's the hardware layer, right? And it might be one site. It might be multiple sites, multiple robots at a site. But they're your data collectors at the end of the day. And so we get the robots set up on site. The next step is the software layer, right? And for that, that's DroneIQ. That is what is ingesting all the data from the robots-- the real-time video-- and essentially enabling that stream both to our client and also back to our SOC, where we're able to do human verification on events and start to build that response layer. And so that's the hardware software side. But then you have the human element, and this is the very practical-- like boots on the ground-- side. And so what we do when we get set up, when we deploy to a site like Maui, we set up standard operating procedures, SOPs. We define different tiers of events with the customer. And that can vary on a site-by-site basis, company-by-company basis. And there's two examples-- tier 2 event could be a break in the fence, a small safety issue-- something that maybe just requires an email or a text message to the local staff to resolve the issue. But then you could also have these tier 1 events. And this is where you need to dispatch a roving officer, potentially escalate to law enforcement or emergency services, depending on the severity of the situation. And so the ability to take the data that's automatically being captured, run that through humans, and then ultimately into procedures that already exist in the physical security footprint with the guards and the local staff is super important. And so I just want to talk through that a little bit. And I'll hand it back to Matt to talk through a couple of real-world examples with our first deployment. - Thanks, Damon. Yeah. So the screen is showing a couple of real-world examples of things, of ways that these Linux helped us right over the last-- almost 10 months now. Last year, Maui had one of the largest and deadliest wildfires in nearly a century. We lost over 100 lives. Thousands of Maui families were displaced because of the damage and homes being destroyed-- billions and billions of dollars of destruction due to the hot, dry conditions that we see here in Hawaii, in Maui, especially on the west and south sides. Brush fires are extremely serious threat to the safety of our residents, of our employees, of our facilities. So this one, I thought, was really timely and a good example. This upper right-hand image, you see, of actual footage from the bear-- I mean, from the drone dog using thermal vision identified a brush fire-- this brush fire is maybe three quarters of a mile, maybe less than that-- from our facility or Kihei facility. And it was started at a homeless camp-- encampment-- roughly almost at midnight. So everybody's sleeping. A lot of people aren't seeing this. The drone dog spotted it. Our SOC per our SOPs immediately called our security operation center. We were able to get the Maui Fire Department dispatched to the fire very quickly, and they were able to extinguish this fire before it could really spread into dangerous proportions. During this whole scenario, the drone dogs took up a post, kept eyes on it, kept us up-to-date with life by the second, coverage of how the fighter-- what the status of the fire was. So it's just really extremely valuable for us to craft our own mitigation plans. Do we need to start evacuating or moving equipment? Do we need to start putting up water cannons? What are our fire prevention strategies? So having that extra set of eyes on the fire really-- I can't even put into words how much it helped us. So it's just incredible. Another example on the bottom right is-- some of the threats that we faced-- as being on a farm, I talked about the crops, the value of our crops-- such as people. It's animals for us-- animals that could come on the farm and really pose a threat to our crops by eating, trampling our crops, whatever that may be. We see everything from pigs to cattle, deer, all kinds of things. So in this example, the drone dog spotted a group of feral pigs that had gotten through a section of our agricultural fencing, and it really headed towards some of our fields-- fields that have high-value crops. I mean, when we're talking high-value-- when you're talking about proprietary hybrids or even proprietary hybrids that are in the making, they could have hundreds of thousands to millions of dollars of future value. And so a family of pigs running through there and eating them could be devastating to us and really set us back. So they saw these animals because they knew it was important to us. There are SOC-- again, called our SOC to immediately identify the threat. We were able to dispatch security officers. We were able to also contact our-- we have a contract with the USDA for wildlife control-- get them in motion to help dispatch this wildlife threat-- so again, that interaction right there. And we had viable product in the field that they potentially saved us hundreds of thousands of dollars. And again, just in closing, these are just two of many examples of how the drone dogs have really increased our overall security posture and become a valued and trusted member of our security team. We rely on these dogs day in and day out. We love them. The team loves them. And we're actually-- we deploy two more in Hawaii-- two more in O'ahu. And we're looking at-- I'm looking at deploying a couple more in California. So I just really thank you for letting me be here and share our story. - Thanks for that. Thank you. Thank you all. We've had a lot of questions coming in. I'm going to start with one. OK. So this was a pre-submitted question as well as a question from Adam. So what kind of challenges do you have with connectivity and communications in rural locations and large campuses? How are you overcoming comms for this kind of application? Damon, you want to take that one? - Sure. Yeah, definitely. So I come out of the aerospace world, and comms is always the issue. I think that's kind of across the board across the industry. So I think the long story short is, we're flexible. The platforms are flexible. Today, we mainly rely on LTE. So we have deals with all the carriers. As part of our onboarding with a customer, we'll actually do an audit to figure out which carrier has the best signals on the patrol paths, and then that's the carrier that we'll go with. So that's our default solution. It works for 99% of our customer sites. We have high-gain antennas-- things like that-- to deal with rural areas. But we have had use cases where LTE just isn't a question. And we have flexibility there. We've done everything from mesh radios to set up a local network when we did security for the Indy 500 this past summer, where we were essentially underground with no comms, and we actually did a mesh network back to a Starlink in order to stream the video-- so again, lots of flexibility there on how we're able to build that network. It just takes a little bit of setup in the beginning to understand the site and the location. - Yeah, great. We had an interesting question. You're talking about how the robot can help distract potential perpetrators-- that sort of thing. What if they attack and damage Spot? And that's always a question that we get. And the answer I always have for that is, that's actually a great example of why it's important to have a robot in between a human responder to a situation like that. And this is a different use case. But in public safety, there's a story on our site about mass state police and a barricaded armed suspect where the robot was lost in that incident. But the end result is the safety of human life. If something does happen to the robot-- it's a very expensive specialized piece of equipment-- that stinks. But it's much better than something happening to a person. And that's-- I don't know if you guys want to add anything there. But yeah, it's a common question. There is-- what other environments have you deployed spot for security? - Sure, yeah. Variety of environments-- we're in 13, 14 different states around the country. As far out as Hawaii now, we've got a cluster in the LA area. We're all the way up in New England. So we've kind of seen everything. I think, particularly in the early days, customers like to put us in their harsh environments first just to see what we could do. So we've done everything from the polar vortexes to summers in Arizona. So it's seen a pretty wide variety at this point. - Awesome. This question comes live. And please use the Q&A button, too, if you have a question. We'll try to get to it. Can you discuss the benefits of your drone dog and aerial drone over just deploying an array of stationary cameras? Damon, you want to take that one? Yeah. - Yeah, I can start. Yeah. I think, at the end of the day, it's a cost perspective. When you're talking about fixed cameras and the amount of space that we're able to cover with the robot, it's just a more cost-effective solution-- when you look at trenching and data and power and everything else that kind of goes into that setup. We're using these robots for-- in the logistics space, you're talking about million-square-foot facilities that we're covering. And so you're not talking about one or two cameras. You're talking about 10s, 20s, 30s cameras. - And I was going to say, we're not also saying, get rid of cameras and just do it this way. - Absolutely not. - This is a whole ecosystem where this fits into that ecosystem. - Yep. Absolutely. Yeah, it's definitely a layered approach there. And the robots have certain advantages, certain disadvantages. I think the ability to have a camera that can move particularly-- again, I'll go back to the logistics space where you have lots of truck traffic, lots of blind spots when you have fixed cameras. Having a robot that can walk down the line of trucks and look between the trucks and the fence line or look between the individual trucks, there's a huge advantages for some of our customers. - And just like the real human guard-- someone-- something might come up on the camera, and now that guard has to go out there with a flashlight and put themselves in a dangerous situation. Or, there could be a much quicker response with spot or with your aerial product, too. Speaking of the aerial product, this is-- someone did ask if you could provide a little bit more detail about what type of drone that you guys use. That came in live. - Yeah, definitely. So we are aerial asset. It's our own proprietary IP. It's actually how we started as a company. We started with aerial asset. Most of us are aerospace engineers by trade. So it's a fully automated drone in a box capable of swapping its own batteries. So it essentially has a house with a robotic arm. So the idea is that the drone can fly. And then once it lands, the robotic arm can swap the fully-charged battery back in about 3 minutes. So if you're interested, we actually are launching our latest drone product at GSX next week. So stay tuned for some more pressers around it. - Oh. How do you ensure that robots are safe from cyber attacks that could either disrupt or take control of the robot? - Sure, yeah. A lot of it just comes down to cybersecurity and compliance. I think one of our advantages is that, because we're a dual-use company and we're compliant to defense regulations on the cyber side, we're able to meet those specifications. So we use encrypted comms particularly when communicating with the robots to ensure that they're all compliant. And then on the back end to the R SOC we meet all the standard cyber compliance required with streaming video data from our customers. - We shared a lot of stats earlier. And we answered this, but maybe it's good to revisit, too. But what's the approximate operating time per robot per day. And this individual also asked, how many robots are needed to cover 8,000 acres? - [INAUDIBLE] - Yeah, I think-- I mean-- so Matt mentions we're doing 112 hours a week for them. And so on average, we're doing a patrol an hour about a half hour, so that the 112 hours a week translates to about 60 to 70 hours of actual roving time with the asset and about 115 to 120 patrols that we're doing. And then I think from the acreage, I mean, maybe Matt can speak to the size of their site and the coverage. - Yeah, absolutely. And I guess I just want to reiterate, the bots aren't the sole solution. It's a part of a layered security approach. So it's hybrid, right? We still have humans running up there. Maybe one day, I'll be out there with 50 robots, but not today. [LAUGHS] So we might-- [LAUGHS] But it's part of-- it's a component. Right now, we're utilizing our bots to really-- they're more centered around the campus areas and areas that have the highest concentration of high-value assets. That'd be equipment, tools, whatever. That's where we're really utilizing them. And we're still utilizing humans and patrol trucks for the more rural-- some of the more rural, harder-to-reach areas-- if that answers your question. - Maybe I can just add a couple of points there. Technically, the robots will last up to 90 minutes on a single charge. But through the best practices and because of working with companies like Asylon whose objective is to provide security services-- not just to sell robots, but to provide that security system in best practice mode-- what they'll do is, they'll determine the level of time that's best used for the patrols and then leave charge in the robot as it's sitting on the dock-- --because more often than not, you might get a signal from a different fixed camera or a door, where you now have to take an emergency dispatch and that robot has charge reserved in it so that robot can get up and go on and be deployed to a situation. So those things come with the experience of working with the customer, determining what's the best practice for that site. So that's very important-- not just the specific specifications of what the robot can and cannot do. - Absolutely. Great. - There was a question-- a couple of questions that are unrelated to security. I'm just going to take them real fast. So are there other use cases in other industries besides from security that your company is involved with? And then I'm going to tie that into another question we got of using the robot for mapping RealityCapture 3D environments-- that sort of mapping. So these are all-- it's one example of some of the other applications and ways that people are using Spot. But similar to security, we do have industrial inspection, which is a really large application where the robot moves around a facility, looking for change in equipment and different types of equipment using thermal cameras and acoustic imagers-- that sort of thing. So if you head to our website, we've got a ton of information about that and other use cases, including government and public safety and more. And there'll be a lot to talk about that 3D mapping as well in the next couple of weeks. So stay tuned for that. What is the operational life of Spot? How many kilometers and how many hours in patrol mode can do it? So again, maybe the bigger question we already talked about-- some of the patrol lengths. But we've got robots in our fleet that are 4 years old. There's a stat that we share that our robot plus our customers' robots have circled the equator, I think, 12 times at this point. And I know, Asylon, you've got your own numbers there, too, of how far your fleet has gone. Yeah, the robot goes, goes, goes. But what happens if there needs to be an intervention or repair-- is a question that came in. Damon, you want to talk about that? - Yeah, definitely. Yeah, I think to Paul's point, the stats of the robot matter less in this scenario because we're a service provider-- security service provider. So in the example of bear, our typical SLA is 72-hour maximum downtime. And so Asylon is the ultimate security provider. We have a fleet of spare robots that we're able to move around the country. We handle maintenance-- everything like that-- so that the customer doesn't have to. And so our goal is, we maintain the SLA that's required by physical security and that presence. Then we handle everything else on the back end. So if a robot goes down and needs repair or maintenance, Asylon is handling that almost invisibly on the back end with Boston Dynamics or ourselves internally to deal with those issues. And then even if it's something small, typically we're able to remotely troubleshoot because we have this 24/7 R SOC that's sitting there. Or, we can rely on techs that will travel around and do the monthly maintenance and things that are required. But again, it's about the service level that's required of security and matching that. And that's the great thing about the service aspect, is the robot, in some ways from that aspect, is invisible to the customer. We want to maintain that. We don't want you to have to deal with it. - Yep. A couple more questions. So Matt, I want to go back to-- this question goes back to something you already talked about. But do you have a comprehensive and concise example of how Spot is more cost effective and efficient than multiple positions filled by security officers? A little more color there. - Yeah. Sure. I can do that. Like you said, I did talk a little bit about the hard numbers behind it. So when we're looking at these dogs, we really have a 1:1 ratio. They're guards. So in our use case scenario-- piloted this-- we were bringing on one dog at first to try it out. He was going to be running 16 hours or two 8-hour shifts. It replaced two guards that would have been working 8-hour shifts. So that is one of how we released our starting point to calculate our savings. But having those dogs there, not having to-- you look at the cost of the dogs versus the cost of what we're paying our folks-- their wages, payroll liability, all of our benefits, incentive bonuses, things of that nature-- and I talked about the fuel savings, which is a real thing here in Hawaii. I mean, fuel is expensive. That's how we came up with that figure. And that's-- I mean, I couldn't-- I don't think I can make it any more clear and concise on how much it saved us. But one other thing I want to point out is that one advantage that you get from having these dogs is you get the active, active engagement from pilots that are walking around and looking for things-- looking for things that you knew-- that they work with you to find out what's important to you in your security organization-- what particular types of threats that are specific to your site in addition to the general threats. And they're dependable. They're going out when they're supposed to be going out. They're patrolling for the time they're supposed to patrol. They're hitting all the areas that we outlined, and they're calling us for anything they see that's out of line or anything that doesn't fit our view of how things should be. And the robots never fail. I mean, they never don't show up to work. They never have to take a lunch break. They never-- you don't have to pay them for overtime-- which is another huge cost savings. You never have to-- they never have to worry about them looking at their phone versus looking at what their situational awareness on the job and what they're supposed to be looking at. So, I mean, the benefits versus having-- we still need humans. Don't get me wrong. I'm not saying we got to replace everybody with a robot. But you can keep the best ones. [CHUCKLES] Keep the best humans with your robots. And like I said, we just tremendously increased our security posture by having this partnership with Asylon. - Amazing. One last question just to close this out, I guess-- how do you adopt Spot successfully for this application? - Yeah, I think-- I'll take a first stab at that. Yeah, and I think you have to be intentional about it. A lot of our setup process starts before the actual deployment. As Matt talked about, we'll build the example patrol routes even when we're in the proposal phase before we have a purchase order. And so we really make sure that there's an understanding of what the robots can do, where they can see that we have an understanding of the customer's needs. And then once we get an order, we actually do a site audit, where we'll go out, we'll validate with the equipment, we'll test comms, we'll test patrol paths, automation, everything like that. And then ultimately begin to ramp up. So typically, we say, the first 30 days-- the first day, day 1 is install. It's out of the box. It's done its first patrol. And then over the first 30 days, it's really about tweaking and tuning the system to ensure that it's properly working with that security team that all the SOPs are established. And then after that, we're rocking and rolling. Depending on the customer, a lot like to do an intro day with the asset maybe with the HR team-- introduce Spot, introduce drone dog, do an employee engagement. Most of our customers do a naming competition for the dog even, so that people can get comfortable with what it's doing and its mission and why it's there. So all those are super important things as well that happen on the back end as we get up to the deployment. - Let me just add to that a little bit. Asylon was extremely professional in helping us roll this out. It can be daunting, especially if you're talking about a tech you're unfamiliar with. But there were professionals from day one. They didn't waste-- they don't waste time. It's one of the things that I appreciated. They didn't just fill you with false promises and say, oh, it has to be perfect. They actually don't tell you they're going to-- you're going to-- they're going to give you the solution until they can actually come to your site and make sure that it's going to work-- make sure that the connectivity is where it needs to be and the terrain and everything is going to work-- because they want you to be successful. And so I really appreciated-- I really, really appreciated that from them. And it's so easy to deploy. I mean, really, from an infrastructure setup standing, all we had to come up with was some level ground and two 110-volt outlets on a dedicated circuit for the drone dog-- for the doghouse. And that's it. It's really, really, truly was white-glove service, and we're very happy with it. - Awesome. Thank you. Well, thank you so much. Thank you. Thank you to all for tuning in. If you are interested, there will be a survey coming with the recording of the webinar. Just raise your hand if you're interested in learning more and connecting with us on this and other solutions. So thank you so much. Thank you all. Thank you, presenters. We'll see you next time. - Thank you. - Yeah, thank you. - Thank you. - Thanks, everybody. - Thanks.