# Drone Autonomy in Power Generation: Setting up for Success and ROI

- Event: NestGen ’23
- Published: March 16, 2023
- Length: 48:09
- Watch: https://nestgen.org/library/watch/Z23mdTjMyqc/

> The transcript of a recorded NestGen session, for reading or for use with an AI tool. It comes from the video's
> captions; automatic captions can mishear names and terms.

## About this session

Attend this session to learn how energy orgs can generate a higher ROI by reducing inspection times and increasing inspection frequencies.

## Transcript

**[0:08]** thank you hello everyone welcome to another very interesting session uh by the way just let us know if you're having a good time how are you finding the session so far tell us in the chat section and also please let us know where you're joining in from and what brought you to this session what are you excited to hear about that way Grant and Kevin can plan their you know uh can uh discuss those points that you asked for today we are going to discuss about a very emerging application where autonomous drone operations generate value or make the most sense that is the power generation industry inspections using autonomous drone systems eliminate the need for specialized personal to be on site in gas facility solar Farms remote oil fields with a simple push of a button energy companies can gather accurate information from the comfort of their offices and to discuss this further I would like to welcome our

**[1:10]** special guest today Kevin todrill from rmos Canada and Grand swastika from DJI North America Kevin is Will Behold a long introduction in coming the general manager of rmus Canada the country's largest commercial U.S reseller he has 20 years of international experience in 15 countries he has an MBA in marketing from ins WC in Paris of post Baccalaureate diploma in Asia Pacific Management from Capilano University in Vancouver and a BA in psychology from Carlton University in Ottawa as the head of Gap Wireless's UA UAV division Kevin was a commercial us Pioneer and a key member of Canada's first DJI Enterprise partner under Kevin's leadership Gap became Canada's largest zija Enterprise partner and a DJI service award winner before being acquired by rmus in 2019. over the years Kevin has built Partnerships

**[2:12]** with leading robotics companies such as DJI Boston Dynamics parrot ortel teal and deep tracker delivering solutions to Canada's largest UAV fields well let's also take a moment to introduce Grant who's currently leading the DJI Enterprise North American Solutions engineering team Grant helps Enterprise users successfully deployed roles before joining DJI he was one of the first to graduate with a degree in the rapidly growing field of unmanned systems Grant expanded on his educational background by starting his own drone Services Company working in the utility field inspecting critical art infrastructure and developing software that uses artificial intelligence to automate the review of inspection photos taken by drone before I let Kevin and Grant share with you how energy companies can generate a higher Roi by reducing inspection times while increasing its frequencies I'd like to share a few housekeeping rules so please type

**[3:14]** in your questions in the Q a tab of the session and it will be answered at the end of the presentation the session is getting recorded and it will be published for viewing soon we also have both rmus and DJI booth at the Expo section so please make sure you visit it during the networking and Booth visit slots and finally do remember to capture your favorite moments of the session and share it on Twitter and Linkedin by using nation23 opendib movement few lucky people might just have a surprise in store and again uh Kevin uh Grant has dropped in a poll so please uh participate in the poll uh finally I would like to thank our title sponsor DJI Enterprise and platinum sponsor voltage Aerospace for joining us in this journey so with that oh over to you Kevin all right thank you everybody I I had no idea my my uh my uh my bio was so

**[4:16]** long it makes me sound very pretentious so I can assure you for those of you that haven't met me and grat you're gonna have to vote for me I I I I probably didn't write that myself so um thank you everybody for joining and uh Craig hello from New Zealand love love the kiwis uh it's uh it's great to have you here from the uh from the other side of the of the world so um I want to I want to get right into things so let me let me just do my little quick uh organizational uh propaganda piece and uh make sure we're over there yes we are look at that so let me tell you about Army West Canada so um we are as we mentioned we are the largest commercial reseller in Canada we have uh offices in uh Toronto just where I am right now and Calgary and Montreal and um we are a member of the global Jordan alliance with uh rmus USA and spirit drones Australia and in Europe uh our partners Nordic unmanned so very excited about that uh We've recently been granted the first uh drone in the Box BB loss sfoc which is

**[5:21]** like a like a waiver uh for autonomous thrown in the Box uh operations in Canada with our partner Ontario power generation so that was pretty exciting after after this we've been doing this for uh pretty much as long as there's a drawn in the Box for about two years we started working with uh perceptive two years ago with the first drone in the box and so we've kind of been doing it as long as it's happened um and then we've had the DJI dock for some pretty extensive cold weather testing and that's part of the autonomy lab which we launched here in 2023 in a partnership with Ontario Tech Mars lab and we're looking to advance autonomous inspections with with drones as well as with spot on the ground so pretty interesting um so I'll take this opportunity to then pass it along to my colleague who comes from a small company in China that you may or may not have heard of and so maybe I'll give you your slide Grant just so you can tell people who you are thanks thanks Kevin and yes the name does often uh pursue me on these sessions for

**[6:25]** folks but my name is Grant hostko with DJI Enterprise um if you haven't heard of us or the leading manufacturer of small unmanned systems started off in 2006 with just 20 employees which has grown to over 14 000 these days with 25 of that Workforce being made up of Engineers and R DS which really encapsulates our product First Vision at the companies have offices across the globe including the Burbank California office that I work out of with our headquarters being in Shenzhen and excited to talk more about DJI doc today which Builds on our other product lines from consumer to agriculture to our Enterprise drones for work perfect thanks Grant so I I just want to talk a little bit about the goal of uh of our presentation today and so grant grant is here to talk a little bit about some of the technical ones one of the things I've noticed having been in this uh in this

**[7:25]** business since 2014 um is there's a lot of you know and we're we're not unique in this in the technology industry but there's a lot of vaporware there's a lot of there's a lot of exaggeration there's a lot of marketing speak that happens and so today I want to talk about you know this beautiful picture this is our goal right which which is what we're talking about which is perfect autonomous drones flying everywhere in peace in harmony collecting all the data we want with no issues perfectly and landing in their boxes and charging and everybody's happy and and this is a little bit about what they're sold and one of the things that we do is I always say we don't we don't make products we make things work at arm us and so really what what I want to do today is is to cut through a little bit of the es which is what we do as a company and talk a little bit about the reality there and what's what's really happening so so I'm going to talk to you specifically about um over the past couple of years of experience we've had with regulators and on the ground with four

**[8:26]** specific case studies uh where this has happened so this is not speculation this is not we think it'll work this is what we did and this is what happened and this is what we learned so that's really our function today and then Grant's also going to share a little bit about uh some of the things that that they've done uh in some of the other applications and then some of the technical uh technical aspects of what we've done so that's that's our goal for today so before I start with the case studies though what I really want to do is I want to reach out to those of you that uh are looking at actually doing this so obviously those of you that aren't oems these are that are particularly end users that are looking at doing this a couple of things you you should really think of and the first one is is really why are you doing this and so our experience has been there's kind of two types of people there's Innovation teams at larger organizations uh or Innovative service companies and so what I would say to Innovation teams is number one partner with an

**[9:26]** operational group at your organization or a potential customer so partner with somebody that already has a use case and is looking to do something but doesn't maybe have the wherewithal and if you're able to do that you'll have a lot of success in the actual uh programs the second group of people is actual operational people and so operational people usually you have great ideas and what I would suggest then is kind of the opposite which is if you have an innovation team at your organization seek them out steal their budget or excuse me use their budget for your own productive uses and what you can offer them is a use case that'll make them look good organizationally so it's really it's a mutual kind of a win-win so just understand and think about what group you are and what you really want to get out of of doing this kind of a program part two is an obvious one what's your what's your Roi going to be so think about how you're going to measure this and and it this will depend largely on if you have to sell this internally or if you're a service provider to a customer what's the

**[10:29]** kind of Roi you're going to promise and there's there's many things besides just it's going to save money or it's going to improve things there's safety issues that are hugely important all kinds of things so no know what it is that you're going to deliver when you're scoping that project look at it you know in three ways is this something can be going to be Greenfield that's never been done if that's the case you want to position it as an experiment so give yourself the flexibility that you don't really know what might happen but take advantage of the fact that it's it's going to be a speculative arrangement if you're going to go to the sort of we're going to do better or do more of something that's when again you want to determine is it something that's going to save me money or make me money or is it going to enable me to be more efficient uh and produce more so that's important one to do and the last part of that is is it going to allow me to do something that I couldn't do before and if that means oh I've never been able to collect that data or I've never been able to collect that

**[11:30]** data that's often that's okay but then understand when you think about it you want to think about am I replacing an existing workflow and if so I better come up with what a better workflow is going to be at least if you don't have it ahead of time be thinking about that as you're as you're actually doing this and the last part of that is think about these things in phases not as separate projects so understand that most of these things are still early stages their beta stages despite the doc it's it's an m30t but it's not an m30t it's not going to pop out of the box work perfectly and solve every problem like a DJI drone will right there's going to be an adoption process there's going to be an experiment process I'm going to talk about this in a learning process so think about it as a phased in approach before you start making Promises of everything being perfect tomorrow I'll talk a little bit more about those phases in a second so I'm going to pass over to Grant now to just uh just talk a little bit about some of the purpose and and what happens

**[12:33]** specifically with with the dock thanks Kevin yeah and with everyone here taking kind of the Liberty that folks understand the idea of the Drone in the Box concept and obviously that's kind of driving the DJI doc from our company's side where we understand that there's a struggle from folks in regards to the labor costs of pilot teams the labor cost of getting pilot teams out onto a location or onto field and then there's also can be high operation skills required to repeatedly collect data so we've obviously seen this grow within the DJI ecosystem and our SDK with the ability to automate inspection routes and repeat them again and again but at the end of the day there is a truck that needs to go out a human there or multiple humans um to execute that operation so the dock is really a new take on remote work and we'll certainly get

**[13:37]** into this more today but with the dot currently we support Waypoint Mission planning the ability to run missions again and again with the ability to utilize it will control coming soon within the Flight Hub 2 and Cloud API for third-party platforms but ultimately just trying to work to make the lives of end users easier and try to address some of the pain points that were here from users in their current operations and we'll go the next slide here Kevin but if you're able to tune into the keynote you might have seen a few of the images in this next slide of the doc uh being tortured as a freedom mentioned uh during the the keynote and I think that's a key part of the system here understanding what the specifications are and making sure that where you want to deploy the dock fits those standards so for example

**[14:41]** we talk about temperature the dock itself can withstand being in temperatures of negative 35 degrees Celsius all the way up to 50 degrees Celsius and then it supports drone flights at negative 20 up to 50 degrees Celsius and really this all circles around the Drone itself as well so it's ip55 protection matches the overall IP protection rating of the system with the core components of the dock being IP67 rated but as Kevin mentioned it's taken years to build a drone system that is reliable enough to meet our users demands and work in step with the dock itself and you can see that the testing field and operations have really been hard at work over the past years developing the system and testing it to make sure it's up to the standard customers expect with a little over 1.5 million flights now so that's a little bit more information behind DJI doc itself and how we've been

**[15:44]** working to make sure it's there to meet our customer standards awesome thanks thanks Ren so I'll get into some of the uh some of the tests that we've uh we've actually done uh in a second but I'd like to just quickly touch on before we get into the case studies on on really this idea of a phased approach and what I mean so again this is this is based on our experiences here in Canada with Transport Canada so so I got to do a little asterisk that the lawyers are saying I'm in no way representing a template or anything that Transport Canada is in any way officially uh announcing this is official Transport Canada policy dot dot dot but having said that what this came about was uh from the use cases that we've actually done in terms of testing and a framework that is going to be something that I think will work for a lot of regulators in some way I will say the phase length is going to really depend on you what you're trying to do where you're trying to do it but this is just a template and so all of these projects we're kind of starting out

**[16:47]** first with the first four areas are going to be a Class D airspace so uncontrolled airspace so start off with um with basically a licensed pilot with a remote that's able to uh to actually take control of the aircraft and who's also acting as a visual Observer and within visual letter sites so this essentially don't really require any specific um any specific licensing it counts as a regular flight so the only difference is you're you're going to still be controlling through the dock but you're going to essentially be ready to take control with the pilot phase two then you're going to remove the pilot you're going to Pilot it through the dock and the software but you're still going to have an observer with a control with return to home functionality and you're still going to function in visual on a site within a kilometer now I will note that anytime your pilot is not physically on site that's still going to count as uh BV loss so in this situation I'm just saying really the V loss is the distance of flight from your Observer

**[17:51]** class three we will remove the Observer and it will be completely done autonomously through the software and the dock with no Observer on site uh class four we will then do that same thing uh Beyond visual line of sites so that will be plus or minus two kilometers and so this is the waiver we have with OPG now I'll talk about that case study in a moment so this is kind of the limit of of of what Transport Canada has allowed uh at this point future developments will be then moving this into controlled airspace and then finally in increased complexity where you'll see the iris automations and you using UTM in some kind of daa avoided systems so again how you do this how long it takes will depend on you your appetite The Regulators appetite and that Reg and that uh and that relationship that you develop but this should be thought of as you're phasing in the the proof of Concepts and this is exactly our approach and how we've done it with these ones ahead of us so

**[18:54]** so with that let's let's let's get in and talk about uh talk about the case studies so uh first case study this is this is a basic one that applies not just to power gen this applies to all infrastructure we've also done this with Telecom infrastructure as well um but essentially this is a site Security application uh in this case the um the asset owner had had a uh vandalism or depending on who you ask a conspiracy kind of attack on some infrastructure I saw some of the other uh case or some of the other sessions earlier talking about Roi how soon is the ROI well in this particular case the you know the Damage Done by the Vandal was in uh the four million dollar range so you can see that if you're able to mitigate those things you're very quickly going to pay off uh the price of of the dock so a couple of things in this one it's a remote area we did have uh we did have internet coverage via a an LTE modem so there was LTE coverage temperatures we were hoping we would

**[19:56]** get super low obviously you can see snow in the background I think I saw somebody from Saudi Arabia in the chat so if you've never if you've never seen snow that's that's what it looks like uh and so uh it was about sort of zero to six degrees so for for that location not particularly cold but um but we were able to run a number of scenarios over the course of a week with uh a lot of success and and primarily what we were looking to prove was um was the repeatability and was it uh was it a function uh that was going to be able to be uh to be done repeatedly so um so what were our takeaways uh we were able to schedule repetitive uh perimeter Patrol missions one of the interesting things about this this was a relatively small location so we did not get close to battery capacity and so that's an interesting thing if you're planning on these things with larger applications you'll have to account for more downtime in between missions so in this case we

**[20:58]** could pretty much do this without a lot of downtime which is important for security but if you're in situations with a bigger perimeter where you can't afford downtime you you may want to look at systems that are going to allow you to dynamically change your batteries as opposed to recharging second part of this um automated change detection can alert you to French breaches movement of objects so that was not something that we deployed on this we did test all the data collection which fortunately passed with flying colors so the next phase we will be integrating uh all of those things with motion detection uh and uh and some AI as well uh existing security systems can trigger drone deployment on alarm again that's that's uh coming soon it was not tested in this one um but the use case has been proven and and the the partner is looking at going ahead with that monitoring Dynamics scenarios in real time again uh as we said uh the dock does not have the

**[22:00]** capability to uh to be flown in real time so we just tested the um the uh um the flight planning uh but with with uh Flight Hub and uh and um the flight based software you'll be able to do this in the near future and we actually did some testing with the flight face offline and we're quite impressed uh with the platform there and then finally uh drill mounted lighting and speakers to deter Intruders obviously with the dock it's based on the m30t uh so there is uh there is lighting that will be available for that or is available for that now um one of the interesting things and you've seen it in some of the other presentations there are also uh third-party docking systems that are able to deploy the m300 and that of course will allow you to use all kinds of different spotlights and speaker systems and that's something that will will be tested on these scenarios in the future and obviously in the end reduced liability avoid repairs and prevent downtime

**[23:01]** and property loss so um so really we see this as uh as being a really core uh core use case no matter what kind of assets uh that you had and we're looking forward to some further testing grab thanks Kevin and yeah I'd shut a little more light on a couple of those and talk about kind of what you can do right now what's coming and what will be supported so we're talking like a third party platform um good point in regards to charging required for the Drone so you can charge from 10 to 90 percent in just about 25 minutes there but if a kind of continuous deployment type strategy or having one on standby is required it would probably be thinking about multiple docs in that type of scenario um when we talk about what the dot can do right now through something like Flight Hub 2 you can see the image on the left where you're able to plan a repeated Waypoint Mission monitor the camera live view both of those are currently possible

**[24:06]** what we're adding here in the coming months in anticipation of quarter two availability for the dock is manual control as well so if you were to see something during the routine flight or need to fly to a certain location you could do so and move the Gimbal and drone itself along with switching the thermal palette so those are kind of to come um you can see currently you can schedule regular patrols as well so you can set it up within Flight Hub 2 in regards to repeat every hour and a half or as needed there um and then Kevin you also touched on kind of the AI side or triggering based off of sensors and that's where third-party platforms NestGen was able I'm sorry Nest gen's the event flight bases FlytNow platform was able to show that during our air Works conference if you saw their their screen recording so that's a great example of leveraging our Cloud API for a more advanced analytics and then you can

**[25:07]** tie that into other sensors as well so I think a good way to think about is Flight Hub 2 provides the base features is that if that is the requirement for the site and then we kind of open up that third-party integration capability for additional items to be built on top and then finally on the PSD gay side and that's certainly something to come on our side you can see I've ever seen the side view of the doctor space for a payload to go on top of the m30 but that isn't currently controllable through the dot controls but still something to be added on our side perfect okay so let's let's move into our second case study and this was uh this was a solar farm uh that was done in uh in Quebec I think I see uh I see some uh uh people from LaBelle Provence have joined uh Stefan so this was um this was a really interesting one because uh this is again one of those things where drones are are pretty proven

**[26:08]** uh Roi use case in you know solar inspections and so really the the application here was not so much to does does drone technology work was can we automate this and and from our perspective this is one of the most exciting ones I think because uh it's it's right for uh AI applications and and really autonomous inspection and one thing if I want to talk about more broadly we're seeing this across aerial ground and underwater robotics is anybody doing asset inspections there's there's less and less maintenance people available and the ones that we have at least here in North America are are aging we're getting a little long in the tooth uh as it were and so one of the things is there's still way too much time spent by maintenance people in you know looking for problems as opposed to doing the maintenance and so that's where autonomy and AI applications are going to have the biggest bang for the buck uh and in a lot of our testing we've been able to take maintenance people on an eight-hour shift that spend five hours of that shift

**[27:11]** looking for stuff and three hours of work to basically six hours of work just from being directed right to these things and we we think that that's going to happen so what we are able to prove in these particular case studies and what we were looking at with the solar farm is are we able to get uh automate detection on modular faults are we able to you know detect vegetation are we able to uh detect um you know string problems and all of those things was was a resounding yes so we were we were able to prove all of those things Grant if you want to just add a couple of things there yeah absolutely Kevin so what we often see in the Solar industry is the need to kind of do that crab walk of the Drone across the panels with the proper gimbal pitch depending on the tilt of those solar panels so that's certainly possible with Flight Hub too right now along with the ability to review both thermal and visual imagery after the fact to help determine if the fault's something where we have a

**[28:14]** crack or vegetation or if it's something internalized that's only going to come up on the thermal camera so once again kind of those bass capabilities available directly through the platform but like Kevin talked about a lot of the value is automating the review of that data and prioritizing issues so certainly can download imagery directly from Flight Hub 2 which the dock can upload it to or utilizing a third-party platforms is what Freda mentioned a bit in our keynote is where folks are seeing some value from deployments in China currently with that ability to pull the imagery directly to the third party platform to the cloud API and then run analysis for teams to make quick decisions in the field and another thing I think we'll you know probably talk about this in the next slide too to get ahead here Kevin is the ability to use the Drone throughout the entire solar project I'm utilizing it you know as a capital expense upon initial setup and being able to

**[29:17]** track the setup of the solar farm itself and use it for that progress both before during and after and with the m30 it's important to note that you're not going to be able to get survey grade results with the half inch wide sensor on there also has a rolling shutter but you certainly can get a good General overview of what's going on at the site and useful information to track that progress and get updates to stakeholders perfect and I think I think I want to touch on that as well Grant I'm glad you brought that up one of the things that we we've actually seen is that there's talk of actually integrating the docs into the actual construction as Grant said and what what I think is exciting about that is you're going to have a built-in in inspection process that is thought about in the design phase and a lot of a lot of what we're starting to see with our infrastructure customers is we're now getting designers coming to us to say how can we build these sites for you know ground robot access

**[30:21]** and then what what can we get with aerial access and how do we we work those two things uh together which is really important so as Grant said you're getting updates on the construction but you've also got a built-in workflow for the continued inspection which is really really important um last thing I would like to add on this particular case study one of the things that was quite interesting was while we were there there was a substation nearby and they and they decided we would expand to do transmission and some substation inspection as this as a test and this is something that you're going to see your projects take on a lot of these things it's like well why don't we just do that since we're there and I would highly encourage you to to do that uh for learning as well as actual results so in this case it was quite fascinating because um we found one key thing about the uh about the substations and in in this use case uh full disclosure it was not actually a DJI drone uh but it had a fixed camera it

**[31:24]** did not have a zoom and one of the biggest pieces of feedback from the asset owner the substation was you need an optical zoom to effectively inspect the substations so in this case which will be different from the one we're going to talk about in a second the high-red mechanical shutter was not important what was important was being able to have a good optical zoom so uh the dock with the m30 is a perfect is a perfect application for those substations and then the second part of that is we're already working on the ground robotics with spot in integrating spot into those substations again for ground robotics that's a really really important use case and what we're seeing now is the integration of ground and aerial autonomous robotics for inspection which to me is the most exciting uh thing that's going to be moving forward so um so I'm looking forward to talking more about that maybe next year by Sally okay so let's keep going here so uh so this is number three and uh

**[32:27]** here we go we're gonna make this video play so this is a wind farm so I wanted you to I wanted you to actually see this so you would actually believe that was really us doing it as you can see there's more snow it was really cold and wind farms have one thing in common they're windy obviously so this was done uh in uh in January of this year uh I I gotta be honest I I don't recommend testing at wind farms in the middle of January uh it's cold the good news is uh the system performed admirably in both the wind and the cold which is nice and again wind farms are a fascinating application because this is probably the most mature market for UAV industrial inspections there's really nobody that owns wind assets that does not see the value of uavs for uh inspection of the blades now the interesting part of this though is you're gonna

**[33:28]** you're gonna ask okay can we do autonomous blade inspections uh well the honest truth is right at this moment uh no you can't will you be able to do it my guess is probably not uh but it's it's possible and part of the reason for that is again this is so efficient you can do a turbine inspection with a pilot for about 300 now it's fast it's high-res it's a little bit delicate um and so it's done very well and very efficiently manually so I'm not sure that there's going to be an Roi necessarily for autonomy uh I'll probably be proven wrong on that but the good news is it totally doesn't matter because even if you're still manually doing your official uh in-depth blade inspections there's a million uses for uh drone in a box autonomous drones on uh on win firms and the first thing with this particular site manager was not a drone person I did not know much about drones but after about two minutes I immediately had 10 use cases and things that we wanted

**[34:32]** to do so uh so we ended up spending a lot of time uh and let's be honest when I say we it was our actually not me it was my team but my team spent a lot of extra time in in the cold and uh and were able to find some interesting things so number one um security applications so these are Big sites remote so we had a lot of poaching in Canada and whenever there's poaching there's people shooting guns so there's security applications uh so that was that was one that was really really important on that um even though it's not the full blade inspection again you can see it's cold and if you can imagine if stuff freezes on a wind wind turbine that stuff can fly 300 meters and you do not want to get hit by a frozen piece of something coming off a wind turbine and so what ends up happening on the sites are they have to suspend uh maintenance in any other operations if they suspect there's ice on the blades so the number one use case or one of the big ones was any time of cold

**[35:33]** weather it's a quick inspection is there ice is it safe for the maintenance teams to go up and again it comes back to what I said before about trying to maximize the team's times and and drone in a box is a great way to do it it totally proven use case there secondly pre-post inspections again if you've been to the assets they're tall it takes a lot to get in and out and so what we did was a pre and a post flight pre-flight to identify exactly what areas of inspections uh that you would you you're going to need to do and that allows you to minimize the time you're actually conducting the inspection and then you're going to do a post inspection there as well so lots of different ones there and then although I promised I would talk about um uh you know only things that have been done I will jump a little bit of ahead on this one this is another scenario where you know they had uh transmission and uh substation assets that we also uh did the inspection on and one of the things I see

**[36:36]** for some of these platforms that are going to be utilizing the m300 will be the use of Corona cameras for Corona discharge and other different specialty sensors so very very exciting for uh wind farms uh so Grant you don't have a slide for this but did you want to add anything to this at all before I move on I think you covered that well Kevin uh just a little time check as well looks like we got seven minutes on the clock still fantastic okay so I'm gonna go uh final one here and then we'll uh we'll wrap it up so this is um this is a damn Spillway so this was this was the first one this is the one where we have the the full BD loss waiver on uh really was done because there was a fatality and not this particular site but at one of the sites where uh there was someone unfortunately lost their life during the the flood uh when the spillway was released um so really this is this is a great application and real it was just looking at can we can we do a quick inspection of the spillway before we release

**[37:39]** um the water and uh and confirm there's nobody there and so what I want to highlight about this one was the use case absolutely worked but there's a couple of things that are important to consider when you're actually deploying this in this particular one we used LTE and so um what happens in the LTE scenario in this one is you're going below grade and we also had trees in this one so rather than flying at about 40 meters that we wanted to fly we had to we had to change the mission parameters to 100 meters and what that did was that dramatically changed the quality of the data we were able to collect so the end result of the first iteration was that we were not able to get the level of resolution we needed to certify that there was nobody in the danger area so so technically for what we wanted to do it wouldn't work in that scenario with with LTE comms and with that particular sensor and so the broader takeaway uh was

**[38:39]** Mission planning considerations you're going to be making you're going to be making changes and that may require an addition or a change of sensor so that's something you're going to find during your testing um you're also going to get into scenarios where you're below grade so that is going to make the comms consideration a child a challenge or potentially something you have to think about is it LTE I heard a session earlier about LTE or is it going to be RF based like the m30t and we know ocusync is the best thing out there there's there's nothing that performs as well so that is going to be important for you to think about it and what's going to happen if you're off off grid and then the last thing will will apply to your sensors is what are your data quality requirements yeah okay so I'm gonna I'm gonna wrap it up uh for me so actually you know what Grant I'm gonna I'm gonna pop ahead to your slide and then I'll go back to that one so I'm going to let you talk about this from practical implementation here and then I'll I'll wrap it up yeah absolutely so you gave the example of a substation or other

**[39:44]** area and just showing one of the capabilities for flight planning currently where you're able to take a automated flight generated Point cloud and software like DJI Terra bring that point cloud in to Flight Hub 2 and then from there based on the camera view you can go ahead and plan your Waypoint Mission so that you could do it on site with the Drone if you need to plan a different Mission or you need imagery captured of a certain area utilizing a point cloud of the site and looking at the camera field of view that's just one of the tools we have available for planning repeated inspections and making sure that you're capturing the data you want to see perfect okay so I'll I'll wrap this up from our experience um really quickly no one size fits all so we were super impressed with the DJI Dock and its integration it's ease

**[40:45]** of use it was very very impressive having said that one of the things is there's a variety of different use cases that are going to require different sensors which is probably going to require different aircraft so I'm sure we're going to have a DJI dock for the m300 and the mavic at some point there are certainly other opportunities already that are there with different sensors but really I don't think that this whole process is going to be a single company owning the technology stack with all due respect to others that are presented I think there's going to be a variety of hardware and software and that's going to be Mission dependent so when you're looking for partners to to do your proof of Concepts and do this you're going to need a partner that's going to look at this holistically in terms of dock aircraft sensors software training and also your um your regulatory engagement Mission planning scheduling remote operations that's going to be very important Communications I talked about you're going to have to consider somebody asked about LTE that's like 4G cellular 5G will be an option soon and then the last thing is

**[41:48]** going to be data hosting and security and data analytics and so your data hosting again I know this is an issue we have in Canada for infrastructure is you need to consider what data you're going to have hosting is it going to be do you need it streamed live can you store it on the aircraft or does it have to go into a cloud if it has to go into a cloud application where is that cloud I can tell you we're working with FlytBase we will have flight based Cloud available in Canada for those of you that are going to do that but depending where you are Europe again know your local laws there are ways to get around it initially for tests but long term you're going to have to have a solution for that and then as far as the analysis we can already do things like thermal anomalies gauge reading that's already done and then the next step which is probably in Alpha later in this year is going to be some kind of decision-making based on the AI so that is that is it for me granted you want to add anything else

**[42:49]** yeah I think you sum that up pretty well Kevin I've tried to cover some of the Q and A's here but certainly feel free to stop by your booth or send me a email or message if we weren't able to cover that today uh okay uh I think we are almost up on time our next sessions already started but uh I think uh thank you so much Kevin this was a very insightful session you went uh you actually uh delved a lot deeper into each and every uh application in use case showing the rois for all of them and how and showing the path to you know deploying autonomous drone systems integrating with the current systems and how it can uh you know scale uh truly we had a lot of questions I however I see that Grant has answered most of them uh I will just check a couple of questions with regards to uh Kevin because he was speaking all the time I'm not sure if he had seen the Q a tab

**[43:49]** so Kevin a few questions are around the regulation so do you think how is Transport Canada you know go taking the autonomous drone adoption is there anything that they are doing to support it uh is there any uh waiver process so if you could just elaborate on that a little yeah so I I would reach out uh directly if you have some specific questions but yeah TC is very in engaged in this I will say they actually we can say autonomous in here but TC is not not big on autonomous the word so I think I think you you want to position it as um as uh repeat operations or something else but yeah they're very active in it and we've we've been working with them uh on some of these applications that I talked about so they're getting more comfortable with um that phased in framework that I talked about so I I think there is a clear pathway to get to where you want to go if you're following some of those basic steps and I think next year we'll see some actual regulations where there'll be a lot more

**[44:50]** clarity in terms of the specific requirements and and the training right a follow-up question to that is does rmus help with regulatory approvals and will they do that for DJ doc so yes to both questions so we're we're working with Javier and the DJI regulatory team and Transport Canada to make that process as simple as possible so all of our implementations of this uh uh in the initial three to six months once we're ready we'll be sort of proof of concept and all that proof of concept work will include all the regulatory support in that uh I thanks Kevin I think one final question and this is from the flight based team and we are asking all our uh speakers uh could you please share your thoughts on open tab approaches like the theme that we have this nation uh where we are talking about bringing in all companies uh that contributes to different layers of the Drone autonomy stack so what do you think are the benefits and how can it help us grow as

**[45:51]** an industry I well first of all I think it's very important because it's it's always I mean DJI makes Incredible products but I think even uh Grant you could probably talk to this but I think the fact that you know they've supported FlytBase and others is they're recognizing with the sdks that that they're not going to be able to do everything and that that's exactly what what we as a as a solution provider want from them and I think it's what the industry wants because we're adding complexity to the process so this is immensely valuable and I thank you on behalf of myself and rmus for putting this together uh and it was really exciting for us to learn about other products and uh and for others to see what what others are doing so yeah congratulations thank you for inviting us thank you so much for your kind words Grand would you like to add to it before we close this yeah absolutely I think you have seen the history from the DJI side and the current implementation with the dock where we have our base platform available so you can complete basic missions and

**[46:56]** processing needs but really where a full solution comes together for a lot of end users is utilizing third-party platforms which can take advantage of the tools available to bring more complete solutions to these specific Industries and build out those solutions that are needed so once again really happy to be here thanks for hosting an event it's been great to connect with so many of you thank thank you so much Grant again thank you Kevin's for the insightful session uh the audience joining us today thank you all for uh you know being a part of this session and you know spending your time out of your busy schedule we have a session that's up right now uh live on stage so please head over that's the last session of uh the day so head over to the stage and check out the the panel conversation uh where we are speaking about what's uh to come in the next decade in terms of B Wellness

**[47:59]** operations and thank you again uh so see you all soon have a great day bye