 I'm Rob Thompson. I've been involved in commercial aviation for 20 years. Also a UAV consultant and analyst. I've done military projects and civilian stuff. Like to introduce Walter Volkman of micro aerial projects. And this topic is on mapping capabilities. So thanks for giving me an opportunity to give you a perspective from a grassroots, grounded kind of a user of UAV technology for quite a while now. So I started a little company called Micro Aerial Projects way back in 2008. It is for profit, mainly making losses to date because of the banning of this activity for so many years. How do I go? How do I go to the next slide? OK. So I could show you lots of stuff that we've done in the past. Our main experience has gained overseas because we couldn't do it here. So as a consultant in cadastral surveying, I had the opportunity to see the world all over the show, especially in the developing scenario. And I see drones as a means to democratize and empower small enterprises. It's not so much just the democratization that's interesting, it's also the empowerment. So drones having become small, light, easy to operate, affordable, very important, transportable, ask those guys who've stood in front of customs officials, field repairable, very, very important, and automatable. Those are the elements that make drones in the developing environment sustainable. If you don't have those characteristics in the technology, you can forget sustainable operation or use of drones in the developing environment. So who are going to be the operators? Hopefully they'll be individuals and small enterprises because they are geographically or spatially more disseminated than big companies and large enterprises like government departments. Who are the people using drones today? It's from the enthusiasts to the professional. So how do we cope with all of that without giving this technology a chance to produce the experience? If we squash it right away, we don't harvest any experiences or any data on which to base decent regulations, rational and reasonable regulations. So Chris Anderson, I don't know whether you know of him, author of a book called The Long Curve, on which I've based my business model, says the drones bring democratization of the aerial view and the aerial view is a very, very interesting and valuable perspective for occupiers of the land. But parallel to that, and very important specifically to survey us is the emergence of a process called structure from motion, which allows you to use off the shelf camera technology to build highly accurate geospatial products. Without that, the drone business would not nearly be as exciting to the geospatial industry as it is today. And it is highly automated. So in this little workflow chart that you see here, the only manual interaction that requires some kind of a geospatial background is the introduction of ground control, if you want to have very precisely georeferenced data, but if you don't want that, if you don't need that, it is highly, highly automatable and basically an easy technology transfer from the manufacturing and the configuration point to the end user. So here's, for example, how easily without you having any interaction and having any personal input, you can restructure an environment to build a very highly, highly realistic virtual world of, for example, the village in Africa. So structure from motion, the structure from motion method empowers individuals and small enterprises to take over the whole mapping process from A to Z, but in small bits and incrementally. So this opens up many, many new opportunities to engage or to use geospatial products. You can produce them as and when needed, be much more responsive to demand than you could be with the conventional way of getting maps out of a centralized, highly centralized enterprise. And the other important thing for those who are in the geospatial business, the accuracy is always the big measure of reliability and all of these things. Accuracy is with this kind of high resolution imagery that we collect as virtually a byproduct of what we do. It doesn't depend on the qualifications of the individuals anymore who are making the maps as to whether the map is accurate or not. The technology takes care of that by itself. So this is another barrier that we can now remove from the mapping environment. And the other thing I would like to just because time is so limited, so in here is we don't always need to produce a map. Often a picture speaks a thousand words. Unless you really have a need to quantify something, if you want to just make an absolute point, you can just take a photograph of it and you can annotate that photograph. And if you've done your flight planning correctly and so on, if ever a need arises in the future to do the mapping, which is the major effort in UAV produced geospatial products, then you can do so and use these annotations that you made, say for example, at the moment of adjudication in a village where there's a first property registration going on. So this mapping, while very, very nice to have, is not essential as far as appropriate application of drones in the developing world is concerned and even in the first world. I think that's all I'd like to take your time on. And thank you very much. Thank you very much, Walter. I'd like to introduce Nina and Georgie Tusev of Tusev Aerials. Thank you so much. Hi, everybody. We're really happy to be here. We'll just introduce ourselves. Oh, Mike, sorry. Hope you can hear me better. We'll introduce ourselves and then we'll show you an example of the work that we're doing right now. So my name is Nina Tusev. And with Georgie, my husband, we started this project called Tusev Aerials. He will tell you, but he's been working in the FPV movement for many, many years. And I come off from an environmental activist background. I've worked at different nonprofit organizations and the United Nations, mostly with indigenous peoples. And so when I learned about his hobby, we realized how powerful of a tool this is for communities and for them to be able to protect their land and their territories, which obviously are facing huge pressures. And I work in the deforestation protection of rainforest fields. So this is a really powerful tool. And thanks to a very innovative and forward-thinking donor here, we were able to pilot this in some countries and a very popular country today, Peru, is where a lot of drone pilot testing takes place. So we'll show you our experience from last summer and what we did there, training communities to use drones and turning the technology over to them. But introduce yourself. Hello, everyone. You introduced me already. So I'll just say hi. And I mean, yeah, I'm from a background of a hobbyist and also a visual artist. I don't know how that's relevant, but somehow processing images makes sense. And building actually stuff made of real materials and foam glue and adding on top of that awareness of some sort of flying control, making work in a real world. That makes me excited and collecting also images and putting them together into some kind of sense. So the video you'll see is an example of the training that we carried out in the northern part of Peru in the Loretto region. And there was a collection of four different indigenous peoples, tribes from three from Peru and one from Panama that gathered together to learn how to use the technology. And Georgie built a drone flying. It was a very traditional kind of a Skywalker type of platform that was outfitted fully for first person view and also to capture high resolution images, obviously, with the goal of making maps and videos later. And the goal was to train the communities to be able to use the drone and then turn it over to them so that it's in their hands to use for their own advocacy. And we were also in an area that had a lot of oil pipes and there had been oil spills. So that was something that we got to monitor as well in the process of the training. So you'll see in the video. In nine days. Yeah. Took nine days. How do we... John. Oh, good. Okay. Community is San Jose de Saramuro, second fort about 20 miles, yeah. So we had to cross the river and approach. It's not pipe serving. We just follow the pipe line because that's the way that only we can find out to reach to the epicenter of the world's drone. And this is, as you can see, this is a map made up of 30 different images. It's just a stitch, yeah. Stitch the images. This is a traditional Amazon song as well. So with this data, the host organization, which is the National Indigenous Peoples Organization of the Amazon Indigenous Peoples, was able to monitor what the deforestation looks like along the line, what the health of the, whether the oil pipe is positioned correctly. I mean, there's a number of environmental indicators that the oil company needs to comply with. So in a matter of this flight, which took 30 minutes, just under 30 minutes, they were able to collect data that is really otherwise very expensive or timely or quite impossible to gather in any other way. Okay, so we'll stop the video here. It's online if anybody wants to, but in order to have time for discussion, thanks for your attention. We'll be back. That's the actual spill there, yeah. Thank you very much, guys. I'd like to have Matthew Lippincott, co-founder of Public Labs, speak next. Thank you for having me here. It's really wonderful to be in this community and on the same stage with such wonderful developers and people who are into flight. I personally have been into flight since I was about four and started making paper airplanes and have engaged starting as a hobbyist and now working with Public Lab. I came into this as a volunteer initiative out of our grassroots mapping program. I was one of the people on the mailing list who hand makes balloons for fun and kites and so I was sending advice and if anyone wants you should ask Shannon about some weird boxes I'd mailed her during the oil spill. And I'd like to, I don't have any slides, but I'd like to talk about two practicalities in mapping. One, because we fly kites and balloons, I think I need to make an impassioned defense of what would be in the jargon called a tethered UAS. Many people when they hear that we fly kites and balloons and they say oh that's really nice but aren't drones going to replace those and I really don't think that will be the case. Washington DC is currently, its airspace is monitored by a captive balloon that is in Maryland, Lockheed Martin tethered blimp and there are many cases in which it's very useful to have a tether, things can't fly away if they have a tether you can get it back and it simplifies a lot of your flight and flight technologies and can actually provide some data security in that you're not necessarily using a wireless link for people who are looking to provide internet, you don't necessarily want internet to move once you've set it up so having a tethered thing in one place is also very useful and simplifying your flight technology allows a lot of community ownership of that technology. Kites are something that can be built and maintained by almost anyone. I personally have a mapping rig that's contributed data to Google Maps where the camera, the kite and the string, in fact the entire workflows is less than $50 so those are things you should think about in your projects that there are cases where these are still very useful technologies. The other practicality I would like to talk about is where we're going in terms of data and storing data and there are some wonderful folks here in the room from the open aerial map program and we've been working with them, as Shannon had mentioned, our software map knitter is for manual map stitching and we've also been working with Steven Mather who's not here at the open drone map program and together we're articulating a concept about how we're using and storing aerial imagery moving forward where we're no longer relying on professional credibility as our only track record of the quality of imagery. We're now going to have millions of people contributing aerial imagery into a system and so the question arises, what does our workflow look like and what does our storage look like and what does data verification look like in the future? And the way we see it, open aerial map will be the repository that we're all going to be sharing of open, publicly available imagery and how we're going to get that imagery there and then verify that imagery, we're going to use structure for motion to process imagery but we also need to store those original photos and so I would say that we all need to think about what's that process going to look like and we at Public Lab would think that and a lot of the people at the open aerial map I believe would agree with us that we should save those individual photos, present those individual photos and have a full record from the original data collection mission to the processed aerial imagery available in the public record and that will prevent some of the issues we're facing with constant surveying around drift of one survey being built on another survey, being built on another survey until you can have an object, the accuracy, the precision of the survey is very good but you're going to have drift from an accurate position in the global and GPS positions and so by saving original imagery, by saving original surveys, it provides the kind of record that we're going to need going forward in order to verify all those systems. Thank you. I'd like to invite my panel to come up, take a seat. Since we're limited on time, I'm only going to pose one question to the panel and then I'm going to turn it over to you guys. The question I have, oh wait till everyone gets mic'd up here, the question I have is who owns this data that you guys are collecting and after it's collected, the public access, should there be any limitations to the people that can access this stuff? And just like start off with Walter. Yeah, I think there's this concept of metadata. You know, if I was the king of the country about five, six years ago, I would have said you can all fly but you have to file your flight plans. To a central database and so that those people who are now engineering all the legislative processes that legitimize drone use can actually have data to build, to have a statistical kind of a base to build regulations on. That would be my kind of thing. Supply the metadata, make the public aware of the existence of the data and then negotiate the access maybe later. But don't hold everything back now just because you haven't clarified the legislative framework in mapping by drones. Sure, and Matthew, what's your take on this? What? Who owns the data? Well it depends on what data you're collecting. I would say you should always try to situate data collection within a decision making process. And so that the people who are collecting data about their own environment have a sense of what that data is going to be used for. And so that limitations can be placed on that collection around community concerns. And I would argue against a policy of always making data public. And say that communities should have some sort of pre-release review of data before it gets put out. Thank you very much. And Nina? In our case it's a complicated issue but it's one that's very important because in our case this kind of data can be used by communities for their own advocacy obviously to map their territories and to seek remedies from the companies that possibly are not following the environmental compliance. So when we do the training we agree with the community, obviously the data is theirs. Especially once we leave there and the drone becomes their own. So we usually can share it with them while we're there for the purposes of assisting with making maps or other, in other ways processing the data but it is for the communities. Excellent. And I'd like to begin taking some questions from the crowd if we have our microphone person here. Anywhere? Sure, if you could please speak loudly and introduce yourself. My name is John Parsons, Noble UAS. I have a question about your viewpoint about data not being for the public. Is that driven off of a royalty concern say a community does a project, they invest the time and effort, their royalties become protected. Is that the main thing that's driving your viewpoint? No, not at all. So I'm sorry to be really quick about it. But I would say it's more about what that data could be used for. In one example I think about a housing and HUD grant that was given in Butte, Montana, a community that I'd worked in. People were looking at the distribution of that money and it was for neighborhood beautification essentially. And that brought up some interesting issues for people because whose house needed to be painted was a reflection of who had money to paint their house. People didn't necessarily want a public record of who was the poorest person in their community who couldn't maintain their house. So people wanted to actually just have town hall meetings that were about the community and figuring out how to distribute that money. Now for accountability's sake, there needed to be some sort of record. So there are public records but they were kept on paper. They're kept in the town hall and they didn't become a part of a published data saying someone says too poor to maintain their facade so we had to subsidize repairing their facade because those were concerns. People didn't want to shame their neighbors in that case. And that's just, I think there are a lot of cases like that and then of course there are cases where the stakes are much higher on how that data could be used. Yes, I can relate that also to agriculture. There's very competitiveness between farmers about who's doing actually well and who's collecting their insurance because of damaged crops. Walter? Yeah, I know of a real life case where a government actually abused the high resolution aerial photography that was delivered in line of a development project funded by the World Bank or the USA idea. I don't know really exactly who it was but a light wing, ultra light aircraft was used to collect high resolution photographs of coastal area that needed some environmental management attention. And in the process of that, imagery was made available to the government which then used this imagery to identify who was transgressing on building regulations and out came the bulldozers. So it was a real big deal. The government actually bulldozed buildings that contravened building regulations. Next question, please. Okay, I'm gonna go ahead and ask her. Go ahead, Tim. So, you know, there's a lot of great, very expensive tools like Pix4D and I'd love to hear about how you think the evolution of free or much less expensive tools are going. And you talked about sort of map knitter, map mill, which are awesome but very labor intensive. Are we gonna get Pix4D like things in the near to medium term future that provide that level of capability for a much more accessible price? Absolutely. The open drone map project is an open source structure from motion package that you can use today to create geo reference point clouds. It is still in early development. You're gonna have to set up a computer pretty much just for it, but it works. And where we see that going is, we've focused on this manual stitching, something that's very easy to use and very easy to teach. In many ways, we've developed some competence with map knitter and map mill around what the interface should be for individual aerial photos. Then open drone map and the structure from motion, those are packages that are dealing with how we're gonna process that imagery. And then open aerial map would be how we're going to package and share and normalize that data and access it. So we're looking at integrating those workflows. We would like to see our aerial image software and map knitter become an interface for dealing with and handling and processing and storing the images themselves and server-side software like open drone map being able to process and serve you a point cloud or process data and then integrate that into open aerial maps so that we have public access satellite data just like Google Maps. For the sake of time, I'm gonna ask the last question and if you guys could give me a 30 seconds, 60 second answer. I would like to know where's the future in data mapping and the future uses for localized data collection, interpretation and implementing this data. The future is in some concerted effort from the development community to drive the initiative to decentralize these mapping capacities. I think that's the future without that sort of dissemination. This mapping technology will remain very much in the hands of the few and will remain a foreign input rather than a domestic local kind of a contribution to development. Thank you. I think the future could head two directions. We could have a backlash against public collection of imagery or a massive expansion of public collection of map imagery. I think it's going to depend exactly on how accessible the technology is and how we situate it within a participatory and decision-making processes. If it's seen as a tool of the powerful against, you know, a tool of the few against the many, then there's going to be a backlash and if it's seen as a tool for everybody, I think we're gonna see a more subtle conversation. Thank you very much, two chefs. I personally think that putting together the fact that technology become cheaper and cheaper with every day more affordable and smaller and lighter and smarter. I see probably exponential distribution of this data and probably becoming instantaneous over the net. Over time, like one more instantaneous availability of the data. Thank you very much. And my one soapbox thing is we all need to pull together and educate everyone and quit the infighting amongst our industry. Thank you very much.