Fly out of the Trough of Disillusionment and up the Slope of Enlightenment

JEREMIAH JOHNSON: All right, everybody. Thank you for joining me. My name is Jeremiah Johnson, and I'm an Imagery Solutions Engineering team lead for Esri. And what I'm going to be talking about today is flying out of the trough of disillusionment and into the slope of enlightenment, which is a lot of words, but we're going to go through it today. What this basically is encompassing is the drone hype curve.

So first, a little bit about me. Like I said, I'm an engineering team lead for the Remote Sensing team here at Esri. I primarily worked with small drones for most of my career. I've worked with them through the home-built days all the way up to our commercial enterprise drones. I started my drone career at 3D Robotics, which some of you might be familiar with, in Berkeley, California.

And there's a picture of me and the drone, and I inserted this picture intentionally because it's kind of foreshadowing a little bit of what I'm going to be talking about. We used to take pictures with our drones like this back when the drone hype was real. I don't think I've taken a picture with a drone in a few years, but this is kind of what we're going to be talking about. It's this roller coaster of the drone hype, where we've been-- I'm sure some of you have experienced this-- where we are today, and where I hope we're going to be going in the near future.

So the big words in the title really refer to the Gartner hype curve, and what is the Gartner hype curve? Well, it's shaped like this with one axis being visibility and the other axis being time. And what this is referring to is when you have a new technology, over time, it gains a lot of visibility, it gains a lot of hype, until it reaches a certain point, which it doesn't. And then, hopefully, the technology advances enough to where we can get back on this plateau. This curve is marked by these kind of milestones.

The first milestone is the technology trigger, the second is the peak of inflated expectations, the trough of disillusionment, slope of enlightenment, and finally, the plateau of productivity. Really long names, you're probably going to hear me trip over them at least once in this presentation, but I think they're really important milestones to talk about. And I'm going to run through each one of these, some in more detail than others, but what I really want to point out is I believe that we are here in the hype curve as far as the drone technology goes. We have been in the trough of disillusionment, and what we're trying to do is get up into the slope of enlightenment so that we can get to the plateau of productivity.

When I take these milestones and apply them specifically to the drone industry, this is kind of how I see them. First, there was the technology trigger, and the technology trigger for small UAS or small drones was the ability to create a flight controller or an autopilot really cheap and really small and really light. And these components were initially created for the smartphone industry and the video game controller industry. So you can thank the iPhone and the Nintendo Wii for really commercializing these really small, really lightweight, and very inexpensive inertial measurement units, compass, GPS units, et cetera. All of these units come together to form a flight controller that we can then use to automate the flight path of a drone.

Well, this got a lot of people excited. This technology validated really excites the imagination, you can think about all the things that a drone can do, and with that excitement comes virtual capital. So VC funds were pouring into this, and what we saw were startup companies that were exploding. Each one were starting their own little niche within this drone industry. Some were making software, some were making autopilots, some are making the drones themselves, but there was a lot of hype going into this, a lot of money and companies getting created through this industry.

Unfortunately, what followed is the trough of disillusionment. We have the reality of technology limitations. Ultimately, the hardware didn't necessarily meet the hype that the imaginary had. And for the drones specifically, we saw government restrictions that actually hampered real world implementation, specifically in the United States and also in Europe. We had governments that were restricting certain flights, maybe it was package delivery or flights over construction sites because they had people on them. And these legal barriers were also, a barrier towards the growth of this technology and actually reaching the hype that it was hoped-- that they were hoped that it would reach.

But what we would like to get to is the slope of enlightenment. Companies find a pattern of drone use that is profitable. You're going to see or we have seen startups dissolve, merge, become acquired. 3D Robotics, the organization that I worked at, our team got acquired by Esri, and that's why I'm here today, is through this acquisition. But we're also seeing the government starts to work with the industry to create guidelines that is more palatable towards the industry and promotes growth of technology. And ultimately, we're trying to get to the plateau of productivity. Drones become boring, productive, but profitable, which is the most important part.

So I'm going to talk about some of these, my favorite milestones. We had the peak of inflated expectations. 2015-2016 was an exciting time. We had venture capital, like I said, that was flowing in. Look at these numbers of some of these companies.

You have Lily Camera that had $34 million in pre-order. I believe that was a Kickstarter. 3D Robotics, which was the organization I worked for, had a Series C that was massive, $64 million. We had promotional videos where we had monkeys that were flying drones. It was insane.

But, in my opinion, what's even crazier is the thought of a venture-funded company, which is Airware, who starts their own venture-funded-- their own venture capital fund using their VC dollars. It was pretty insane. So you had Airware that was using their borrowed dollars to borrow to other companies that were developing flight software. It's pretty incredible.

And then we saw our first successful drone delivery made in the United States. I mean, this was in 2015 that this happened, but we really didn't see true adoption of some of these technologies even today.

Well, what followed the peak of inflated expectations? We have the trough of disillusionment. Reality sets in.

We see Forbes and TechCrunch and Engadget articles kind of writing about this fall of the drone industry, 3D Robotics having to dump everything and get out of the hardware market. You have Airware that crashes and shut down after going through $118 million in VC funds. Even DJI, which is a household drone name that's still the highest or the most-- they make the most drones in the world, they're having to make layoffs and cuts to their workforce as reality sets in. And Amazon Drone Delivery, the Amazon Prime drones that everybody was really excited about, they were supposed to launch deliveries in 2018. We're still not seeing that today, other than small proof of concepts.

Well, what's after the trough of disillusion is a slope of enlightenment. This is where we're trying to get to. What does this slope look like? What are some of the characteristics of this slope?

Well, one of the first characteristics is confidence in the product, confidence in the technology. Obviously, this is a time when you need your customers, your users to trust in the products to be able to use it in a variety of different ways. The technology needs to bring high levels of productivity. This is no longer a science experiment. It's not research and development anymore. If we really want to put this into the hands of our workforce, it needs to be very productive and be better than the technology before it.

We also see with the success with the confidence, more organizations want to be involved. So organizations who might not have been technology-forward, maybe they're waiting on technology to kind of settle down before adoption, this is when a lot of these companies start to get in. They see other organizations be successful. They kind of pave the way for productivity and profitability using technology, and they start to adopt it. And finally, it's easier to make forward progress. As we have a higher number of users using a specific technology, it is easier to make forward progress, not only because of the financial reasons, but also, you have a lot more use cases and a lot more people kind of paving the way for workflows.

I want to bring forward a case study that, in my opinion, shows what this slope of indictment looks like, and it is a-- it's a customer of ours, Skanska, specifically the Skanska Norway office. And they have a fine example that I'm going to be sharing of how they use drones and specifically, an Esri product site scan, to automate a lot of their drone inspection and mapping workflows on a pretty incredible project, the Nordeyvegen Project.

Just a little bit about the Skanska Norway Drone program. so Skanska is a global company, but I'm specifically talking about their Norway team. They have 88 drone pilots and 788 total users of the software. To date, they perform-- on this project, I believe, they've performed 740 flights over 86 specific projects. You'll see as I get into this Nordeyvegen Project, it's pretty large. And then, 40 of these have been completed.

So what's the Nordeyvegen Project? Well, it's this incredible project to connect five Norwegian islands to the mainland. These are islands that have never been connected before in the previous-- they've been primarily ferry-driven, but now the country is putting millions, if not billions, of investment into connecting these through a series of bridges and tunnels. What this consists of is 2 and 1/2 kilometers of ocean fill, three bridges, 13 kilometers of tunnel, three of these being underwater, which is incredible to think about, totaling 35 kilometers of a stretch of roadway that connects these five islands together.

Well, what are the benefits of this project? Look at this. Just in Google Earth, it looks pretty incredible, but where the drone technology really starts to come into play is specifically site visits over the course of construction. Like I said, these islands have never been connected before, it's all ferry driven, so even during the construction, even getting construction progress, is actually pretty difficult. You have five islands. Some of them, it takes 10 hours plus to reach.

You have six artificial islands that are accessible only by boat. You can see some of these, look here. And this is one way-- drones are one way to bring this reality into the office because the head office is not stationed on any one of these islands. The head office is somewhere else. And these live updates almost of the drone progress gets pulled in and shown through stakeholders, not only on the construction site, but in the office itself-- in the main office.

Some of the project benefits for drones specifically is its ability to measure difficult areas. So there are areas that are difficult or impossible to measure using traditional methods. This is a screenshot of measuring a point cloud that was derived from a drone, and these measurements can be done in both 2D and 3D. The site scan automated flights allow Skanska to collect more data because it's automated. And because it's accurate, they can trust this data, much to one of the qualifications of the slope of enlightenment.

And then, using these point clouds and using these orthomosaics, they're able to QA and QC their work. This was something that potentially wasn't even done using traditional methods, and now they're able to QA and QC a lot more efficiently and QA and QC things that they might not have been able to in the past using drones.

Some other benefits is some of the project engineers, those who are on site, are able to pull in the imagery from the drone data, and they're able to overlay some of their CAD data on top of it. So you can see here on this screenshot, you can pull in some of the design files and drape it on top of the drone data so that you can QA and QC almost in real time, seeing the as-built versus as-designed.

Back over at the head office, they're using this data for data management. Because everything is stored in the cloud, so Site Scan processes all of the data in the cloud so that's where all these orthomosaics and these point clouds are stored, it simplifies their data management. And allows everybody in the organization to have access to the same 2D and 3D data sets.

But what I found was interesting is you have this client relationship piece. As you might expect, connecting these five islands is a pretty important project to the community, but it's also something that maybe not everybody is super excited about, seeing large amounts of construction and tunneling and boring into islands that them and their families have been living on for years, if not hundreds of years. So what Skanska has done is build out a client relationship portal where data from Site Scan is shared on a publicly accessible portal so that the locals can see in nearly real time, the status of the project, kind of what they're doing, and kind of start to build these relationship links between Skanska and the locals that live there.

What does this data management strategy look like? I have a workflow here that runs left to right. So after the drones are flown, for this project specifically, it's a Phantom 4 RTK drone that they're using, and they have an iPad app that connects directly to that drone. And it automates their flight so the drone goes out and collects a series of images. Once the drone lands, those series of images are uploaded to Site Scan Manager, and that's where this workflow starts.

Site Scan Manager processes those images using photogrammetry to create these 2D orthomosaics or digital elevation models, as well as 3D data sets like, a 3D point cloud or a 3D mesh. These data sets can be viewed, measured, analyzed right inside of Site Scan Manager, as some of these screenshots have shown, but ultimately, what you want to do is pull these data products from Site Scan into some of the tools that they use every day.

So this web viewer is a web viewer that is used by not only the stakeholders, but also that customer client relationship side, where they can view the most up-to-date orthomosaic or this overhead view of the project and see where it's at today, but they can also pull in the point cloud into software like Gemini or Civil 3D. Civil 3D, in particular, can pull this data natively through a connection with BIM 360. I'm going to talk about that there in a second, but this point cloud can be pulled straight into Civil 3D, and that's where a lot of these design files or these BIM files can be pulled in as well to measure the as-built versus as-designed.

BIM 360, it's Autodesk's file management or project management dashboard. All of these data products can be pushed straight from Site Scan Manager through BIM 360 through a Forge connection. So any of these data sets can be published using one button straight to BIM 360, and once it's in BIM 360, these data products can be used throughout the rest of the Autodesk suite. So Skanska specifically has BIM 360 to distribute and view the 3D mesh and the 3D point cloud.

And in ArcGIS, which is Esri's applications, so ArcGIS Online is a web application that is similar to BIM 360, except instead of leaning more on the CAD side, it's leaning more on GIS. So what the team over at Skanska can do is pull in the point cloud and the orthomosaic, put it into ArcGIS Online, and see their construction data in relation to their GIS data, their terrain data, and really, environmental data for the rest of the project, which are typically housed within GIS.

Moving forward. So the Skanska team, like I said, they're very forward-thinking when it comes to technology. They've really started to standardize these drone workflows, but there's still a few pieces that they're working towards, and I wanted to take some time to share those with you here. So today, the team members depend on desktop applications. These Autodesk applications that run on a desktop computer usually requires a large desktop computer in order to run and visualize them, but tomorrow, what they really want to do is publish both their Autodesk design files and the drone data files into ArcGIS Online so they can view this 2D and 3D data in a web browser.

So something I didn't mention before, which I'll briefly mention here, is that through an alliance partnership between Esri and Autodesk, we have kind of a connector between the GIS world and the CAD world, so data can flow to and from. So a lot of these 3D design files that you see here can actually be pushed into ArcGIS Online, and a lot of the GIS data in ArcGIS Online can be pushed into the Autodesk suite of products.

Here's what this looks like. So what we're looking at is a 3D scene inside of ArcGIS Online. And here in the 3D scene, we can see our BIM files for this bridge in particular. And on top of that BIM file, we see the drone data. In this case, it's the 3D point cloud that is being shown on top of the 3D BIM file.

And through a Site Scan to ArcGIS Online connector, this is really easy, it's a one button push, but what we're still working on is these BIM models to ArcGIS Online. Think about how complex some of these BIM models can be. Autodesk has done a great job of being able to take every item of a construction site down to the steel bridge bolts, is an example on this slide, and have those in as a 3D model. Well, when you pull those 3D models into ArcGIS Online and try to view them in a web browser, it kind of blows the system up. And you don't necessarily need all of that information, all of that data inside of ArcGIS Online because the intent is to show really, just the project status of the BIM files today as-designed and then where the construction progress currently is using the drone data. So that's something that they're working on optimizing.

Here's a great example of some of the power of using this 3D scene. So this is another view of the 3D scene that we were just looking at, of this bridge, and now showing what we can do with some web-based analysis. So here, we've taken a slice through. We can see inside of the BIM design file here, as well as a slice through of the drone data. And remember, this is all happening in a web browser. So these are Esri and Autodesk data products coming together in the web where people can analyze them using their favorite web browser.

So you could do this anywhere in the world with an internet connection. You could do this on an iPad if you wanted to do. You can really sort of scale this analysis to where it's not just the engineers with the powerful computer on the construction site that can do some of these analysis. Maybe more stakeholders around the world can start to do this now that the data is available in the cloud.

Within the scene viewer, you can measure height, length, and area. And then, of course, right here, we see the Section tool, which is used to slice through the data.

That's, more or less, the Skanska project that I wanted to introduce to you, but I think it's a great example of how an organization that is technology-forward, but also has real projects that they're working on. They can't just do an R&D project or a science project and try to get something working. They have to be able to trust the technology and use the technology in a profitable way, and this project kind of proves this workflow, and proves out what they're doing in Norway on this pretty massive project.

A couple of slides to close us out. Let's fly out of this trough of disillusionment. We've had a wild ride in the drone industry, but the market and our industry is showing that there are practical, productive, and profitable uses for this technology. In order to advance into the slope of enlightenment, which is the next stage after the trough, it is key to move past the R&D stage, the science project stage, and develop standard practices for adoption in the organization.

Skanska has done this. They've scaled their drone program out across their 88 pilots, and they have data pipelines and data workflows that are flowing from Esri and Autodesk software's into portals, dashboards, and experiences. They've done this. They've kind of paved the way and shown what is possible here. Overall, Skanska is a great example of how an organization that leans forward into innovation can standardize a way to use drones in their organization. And I hope that some of the learnings that they have and some of the kind of cool features that you've just seen inspires you to develop standard operating procedures for your organization.

Well, what's next? Well, let's make drones boring. After the slope of enlightenment, after we have organizations that are really paving the way and showing how drone technology can really be used, so we're stripping away the hype, we're starting to think in real business dollars and productivity, next is the plateau of productivity. Unfortunately, this means that drones should be boring.

Think of the most common drone or robot today. They tend to be water sprinklers, Roombas, things you don't really think about as being innovative anymore just because they've been in our lives for so long. That's where these small UASs or drones are going to be headed given enough time and cycles in our organization.

When you make drones boring, you make them standard, you make them reliable, you make them trustworthy, and ultimately, this brings higher profits and productivity and greater technological advancement as the user base grows. I know that I showed a construction example, but construction isn't the only area where drones are starting to see a lot of enhancements. We see it in infrastructure or inspection. The FAA has infrastructure dollars that are specifically for drone inspections of their aging assets.

So let's work to make some standard operating procedures and some standard workflows for the drones in the industries that we work on so that ultimately, we can make them boring, and thus, bring more productivity to drones because they certainly do deserve it. They have brought immense value to the organizations that have been using it, and I know that there's a lot more there.

That sums up the talk I have-- I wanted to talk about today. I am very active on LinkedIn. So if there are any questions or any articles or your own experience riding the hype curve-- the drone hype curve, as it will, I'm available on LinkedIn. You can hit me up, and I'd be happy to-- more than happy to connect with you there. Thank you.