Speaker 0 00:00:00 So at Indoor Robotics, we develop autonomous indoor drones for safety and security.
Speaker 2 00:00:15 Hello everyone and welcome to the podcast. My name's Jim Beretta, and thank you for joining us. And thank you for subscribing. Uh, I'm excited to have Emmett Moran, who is the co-founder and CTO O of indoor robotics that develop autonomous indoor drones for safety and security monitoring missions. Emmett holds a bachelor and master's degree in software engineering from Insa d Leon in France. Prior to founding indoor robotics, he served as VP technology innovation at a startup company called Timmy, which develops a personal assistance robot where he focused on the developer's experience and prototyped new capabilities. Before Timmy Emmett worked at Intel in the perceptual computing or the real sense department in the innovation and advanced technologies group. During this time, Emmett established a research and development team in the field of robotics and computer vision, where he led the development of software and hardware products for computer vision and for robotics. Emmett, welcome to the podcast. I'm glad you could join us. Thanks,
Speaker 0 00:01:14 Zim. Very happy to be here.
Speaker 2 00:01:15 Hey, can you describe your drone, tell our audience since we we're not video right now, uh, what does it look like? And you have a unique mounting device.
Speaker 0 00:01:24 So our drone is, um, is about 30 and 30 centimeters. Something like, um, like a Roomba, if you will, at the size. Um, it's, uh, fully enclosed around, uh, for, uh, for safety. I would have a prop guard around it. It is mount on the ceiling. So this is our docking station. We call it the tile. It's like a ceiling tile. One of our innovations that we developed, the idea is that drone is docked on the tile and it's constantly monitoring the environment this way. So also when it's docked it, uh, it provides valve.
Speaker 2 00:02:01 And for its stalking, I'm assuming it's charging when it needs to charge.
Speaker 0 00:02:05 Yes, absolutely. So the docking station is also a charging station. So while it's dock, it's charging.
Speaker 2 00:02:11 So how big is the security industry? I mean, we talk a lot about robotics and I, robots are DR and drones are robots and vice versa. So how big, uh, is the opportunity for you?
Speaker 0 00:02:20 Well, the opportunity is pretty big. We talk about something like, uh, an industry of 250 billion. Um, cause we are something in between the security and smart building industry. And, um, you know, if you take only, um, I think the largest security firm, uh, it's Allied Universal. It has more than 1 billion guards around the world. So it's, yeah, it's pretty big.
Speaker 2 00:02:52 And so when is your drone used?
Speaker 0 00:02:55 So the drone is, it's fully autonomous. I mean, so while it's docked, I I explained before, but, um, the idea is that it's scheduled for specific missions during the day or night or when an alarm triggers it, it can detach, perform a emission, uh, detect hazards people and, uh, go back to its, uh, docking station without the need of anyone flying it. So it uses, it's being used in, uh, security, um, security settings for guarding, for example, can be office, can be, um, different scenarios like retail. Uh, it can be used for safety inspections, uh, in warehouses, in the data center has been used for inspection of the actual, uh, servers, for example. So it's a monitoring platform basically.
Speaker 2 00:03:48 So what's an alarm, maybe like a temperature alarm or, or something like that. How does, what's that look like,
Speaker 0 00:03:53 You think The best, uh, scenario you can think of, it's like in an office scenario. Uh, when you have a security alarm, so a sensor, motion sensor or a door has been opened, the alarm triggers an alarm, the alarm system triggers an alarm. Our system is integrated to this, uh, VMs or alarm system and can send automatically handle to inspect to see what's going on. Many times it's false alarms. So instead of sending someone a physical person to see that it's just a false alarm cat or something like that, uh, we can actually send tander to inspect and save a lot of money at time.
Speaker 2 00:04:32 Yeah, it must be, like you say, a lot of, lot of alarms. You could imagine being in a million square foot building and all of a sudden you've gotta send somebody, it could take them half an hour to get there.
Speaker 0 00:04:41 Absolutely. So it's gonna can get, we can get there very fast and obviously the quality is high because everything is recorded. It's not just someone getting there and saying, ah, it's nothing. Um, and this is a very, it's a very dull task for most people to go and to inspect if it is an alarm triggered or just doing a patrols. So this is where, I mean, I don't need to explain it to you and I guess the <laugh> I guess the listeners as well. But you know what, what robot are really good at is they're doing this is dull, uh, tasks, doing it again and again and again and not getting bored.
Speaker 2 00:05:18 And I guess too, uh, you're there within seconds, uh, and a security guard is there within maybe 20 minutes or something. Uh, it everything becomes more timely.
Speaker 0 00:05:27 Absolutely. So we can be there very fast, probably the fastest, and record everything and then we can act, act actually afterwards and investigate what happened.
Speaker 2 00:05:40 And so how do you sell your service? Is it like a robots as a service model or do you use distributors?
Speaker 0 00:05:45 It is a robots as a service. We provide the platform. We use distributors as channel partners. For example, in, uh, north America. Uh, we in a, in the USA we use, uh, a d t. We're working with AD t. We partner with them, uh, in other parts of the world. We are partnering with different, with different channel partners. And uh, the model itself is robot as a service. So it's an annual subscription, uh, for the, for the service.
Speaker 2 00:06:12 And is there a particular environment that works best? Like you've mentioned warehouses, you've mentioned, uh, like server farms and data centers and that kind of thing.
Speaker 0 00:06:20 Right. So, um, we are looking at those four verticals. So office spaces is one. Uh, we have several customers already. Uh, having those, enhancing their guarding services here in Israel. We have, uh, not sure I can say the the name, but it's one of the biggest tech companies in the world. We have a full site here that is been patrolled by, uh, by tens. Um, so this is one vertical for mostly for security and safety after hours patrols, uh, retail the same. We look at data centers. As I said, it works, uh, pretty good. Cause there, it's interesting because many times you don't want actually people to go inside the holes themselves, right? There's a lot of security around that cyber security and actual physical security for, uh, for the servers. Uh, so they, a lot of, uh, our customers prefer this, uh, this automated, uh, solution.
Speaker 0 00:07:19 So in warehouse, uh, in warehouses, uh, we are specifically, uh, interested, uh, and, and our customers interested in us for safety checks. Uh, we have a partnership with, uh, with Nestle. Uh, we have several, uh, warehouses. We are where we are patrolling the warehouse, looking at the shelves and making sure that the shelves are safe because, you know, there are the forklift workers who are coming and moving all the time and big heavy equipment. And they, sometimes they heat the shelves and it can cause some of the, some of the boxes to fold eventually, or even to shelves. So we can actually do a routine patrol and making sure that everything is safe for the workers. And it's something we are also very interested in.
Speaker 2 00:08:10 I you made a really interesting comment just before about drones being a augmentation to existing security, right? Like you're not replacing security personnel.
Speaker 0 00:08:19 Right. So as we see ourselves, we are somewhere between a security guard and a security camera. A security camera has advantage. It's very, uh, it's stationary and it's there 24 7, uh, filming. And there is a guard that actually can manipulate thing and it's smart, it can actually take decisions and, and do things. Um, and we are somewhere in between, we're actually augmenting the guards, committing the guarding service to have more quality, better patrols, more patrols. Cause we are not getting tired. I mean, we as robots <laugh>, I'm not getting tired. Our, our robots are not getting tired. Uh, they're filming everything. We have a 360 camera on the drone itself. Uh, didn't say before. We have a thermal camera. Uh, so we can actually, we have also other cameras looking all all around and other, uh, sensors to understand the environment. And uh, we are recording all of that.
Speaker 2 00:09:18 So Emmett, how is the navigation performed? You must have some tricky environments or obstacles that you have to, uh, be aware of.
Speaker 0 00:09:25 Yeah, um, I think navigation is one of our biggest ips of what we are, um, working quite a lot. Um, we are using several technologies in order to, to do that more indoor navigation. We don't have gps, so we have to understand the environment ourself. We are using, uh, what's called inside out technology. We don't place, uh, external, uh, cameras that look at the drone and tell where it is. It's the drone understand its environment. We're using several cameras on the drone to understand where it is. We're using also, um, um, distance sensors around to also understand and avoid obstacles. So different technologies, we kind of, uh, call it the Swiss cheese model. Cause in each technology, in each, uh, algorithm stack, uh, there are holes and we're using different stacks like that. For example, you can imagine, um, glass wall. Glass wall is something very difficult for specific sensors like cameras. Other sensors can actually sense it better. So this is one example of how this kind of a Swiss model approach can help us avoid obstacles.
Speaker 2 00:10:43 So you mentioned a couple of different sensors that you're using. We've got a couple 360 cameras. Um, are there other, uh, other sensing technologies that you're using? Or have you kind of have We covered them all.
Speaker 0 00:10:52 We basically covered them all. We have different kind of cameras. So it's, we have, so it's a bit different kind of camera. We have, uh, it's kind of lighter, but it's not, uh, it's a static lighter around the ca around the, around the drone. And uh, obviously the other inertial management units and stuff that is more, uh, low level. But it's actually pretty useful.
Speaker 2 00:11:14 So I was gonna ask you a little bit about, I know we've kind of covered some of your Bullseye customers and they're these people who are operating really large, um, uh, areas and have to have security and have to have p maybe places where they don't want people. Are there any other bullseye customers that are kind of interesting to you?
Speaker 0 00:11:31 Um, I think generally speaking, enterprise, uh, customers that are interesting cause they have usually several, uh, use cases. They have office, they have, uh, um, some kind of a warehouse data center. And they are specifically interested in having the unified approach, which is high quality, can actually save the data, analyze it, uh, across their, uh, their facilities
Speaker 2 00:12:05 I guess. And that's one of the things that they get with, uh, indoor robotics, right? Is they get data that they've almost never had before. Yes, they have static cameras, but now they can actually go right to the, uh, to shelf number seven and, and have a look at it on video.
Speaker 0 00:12:20 Absolutely. This is it specifically interesting because you have this fusion between visual thermal, other data together with this location. And it's not just placed statically in the, in the day on the day that you installed the camera or the sensor. It is dynamic. Uh, for example, you can decide to do to specific, uh, door to check it or specific machine that you want to verify. And tomorrow there's another machine or you want to look at the windows, suddenly you can all the time change the emissions and you get the data historically depends on your, on your log, uh, retention policy. Uh, but then, uh, you can actually get a lot of information across the time, get trends, how things are getting hotter or colder with time. We are, we can detect thermal leaks from windows, stuff like that. So it's something that is, you get only from a machine or a drone in our case that is moving around and filming and capturing data.
Speaker 2 00:13:22 Yeah. And I guess a thermal, uh, leak of some sort could just be a poorly installed window or could be an open window. Right?
Speaker 0 00:13:28 Right. So it, it's actually a very interesting thing cause we're developing right now our anomaly. So part of the thing is basically collecting videos and um, and thermal and just collecting the data, but we have also our AI on top of it, our classification. So some of the stuff we are doing are more, uh, normal like people detection or uh, movement motion. Uh, but some staff are doing more, uh, uh, specific for our use case. For example, open doors or open windows, uh, together with its location. And we can detect if a door is supposed to be closed, but it's actually open or the other way around. For example, a fire door. It's supposed to be open all the time, it's actually closed. And uh, we can alert if something like that, uh, is happening.
Speaker 2 00:14:15 And are there any other customer use cases? I, I suppose too, you probably have success stories, but probably a lot of your customers aren't telling you much about their successes.
Speaker 0 00:14:25 Uh, yeah. So we have, so most of the time it's, you know, most of the time the drone is there to, to patrol, making sure that everything is okay. And most of the time things are okay and this is the, this is the interest, this is why it's a very dull, uh, work to do. Yeah. Um, but we have detected, um, electric scooter that someone left inside the, one of the offices, uh, plugged and it could, uh, it could, uh, it's it's against the relation and it could cause fire where we know there a lot of stories, how could this, uh, happen.
Speaker 2 00:14:59 Yep.
Speaker 0 00:15:00 And, uh, and we have other staff that, uh, some of the customers are not all of them telling us we know in the warehouse because it's, uh, it's a poc. We did detect it, uh, it was a p piercing. So we, we did detect some problems in the shelves that, uh, we helped mm-hmm. <affirmative>, uh, to identify. So, no, it's really interesting to see how we can actually detect those anomalies. Sometimes they will never know about it unless we are there and can, we can actually save a lot of money and, and lives.
Speaker 2 00:15:28 Emmett what does the return on investment look like for some of your clients?
Speaker 0 00:15:31 So it really depends. Cause some, some clients they look at it as a way to save, uh, to save on personnel in some case. I mean, they not reduce totally, but uh, but they reduce some of it. For example, um, one of the cases they talked about in the office, we managed to reduce something like 60 to 60 to 70% of the cost of guarding. Um, in other cases, like in the warehouse, it's not about saving the money, it's increasing the safety. They, they, they're not firing people or they're not, uh, causing any changes in, in their event power to just increasing the quality that those checks were done before by personnel in a very scars, uh, cases. And now it is repeatedly done by drones
Speaker 2 00:16:23 And personnel can be used for other more important, uh, situations. Right. When they have drones installed,
Speaker 0 00:16:29 Absolutely they can actually take decision if they see an anomaly and they can, they are much more efficient when they need to take decisions on top of the filmed, uh, data, they can actually understand what's going on, they can share it with other people. Um, you can use the personnel to do tasks like monitoring, uh, a larger environment to understand what's going on instead of just one area that you said, okay, this is the most important and this is what I
Speaker 2 00:16:59 Can do. So I suspect that a lot of your customers are in growth mode, right? Because they have larger facilities all of a sudden. And this might be an o o opportunity for them to use a new innovative tool.
Speaker 0 00:17:09 I think that, uh, oh, it depends. Some of them are, are, yes in growth, some of them are, it's in their normal operations and they just want to, to take it, uh, to take it to the next level in getting better quality.
Speaker 2 00:17:22 I wanted to ask you, when you're not programming drones and uh, selling drones, what do you like to do?
Speaker 0 00:17:27 Surfing and, uh, something, you know, uh, and best in Tel Aviv. We have a good, uh, uh, good, uh, sea over here. Uh, good beach. So like surfing and uh, and program some more stuff. <laugh>, <laugh>,
Speaker 2 00:17:43 You know, I forgot to ask you, but are there any future technologies that you're working on that we can talk about?
Speaker 0 00:17:48 I can can only say that indoor robotics stand is our first, first product. We see the indoor space as space that can do, we can do a lot more with automations. Um, and we are looking at, the first step is about understanding the, the space. This is why Tando is the first product. It's about understanding, collecting data, understanding what's going on. And the second step is about create, making more impact by manipulating, uh, the environment.
Speaker 2 00:18:22 And thanks for participating today. And how can people get ahold of you?
Speaker 0 00:18:26 Eh, I think the best is by, uh, LinkedIn. Just look at, uh, Amit Moran, a m i t m o r a
Speaker 2 00:18:33 N and I'll put that in the show notes as well. Thanks, Amit.
Speaker 0 00:18:36 Thank you very much, Jim.
Speaker 2 00:18:38 Our sponsor for this episode is Earhart Automation Systems. Earhart builds and commissions turnkey solutions for their worldwide clients. With over 80 years of precision manufacturing, they understand the complex world robotics, automated manufacturing and project management delivering world-class custom automation on time and on budget. Contact one of their sales engineers to see what Earhart can build for
[email protected]. And Earhart is hard to spell. It's E H R H A R D T. I'd like to acknowledge a three, the Association for Advancing Automation. They're the leading automation trade associations for robotics, vision and imaging, motion control and motors, and the industrial artificial intelligence technologies. Visit automate.org to learn more. Painted robot builds and integrates digital solutions. They're a web development firm that offers SEO and digital social marketing can set up and connect CRM and other e r p tools to unify marketing, sales, and operation. And you can find
[email protected]. And if you'd like to get in touch with us at the Robot Industry Podcast, you can find me, Jim Beretta on LinkedIn. We'll see you next time. Thanks for listening. Be safe out there. Today's podcast was produced by Customer Attraction Industrial Marketing, and I'd like to recognize my nephew, Chris Gray, for the music, Jeff Bremner for our audio production, my business partner Janet, and our sponsor, Airhart Automation Systems.
