00:00.00 Paul All right. 00:00.31 archpodnet Um, welcome back to the archeotech podcast episode one seventy eight and I hope over the commercial break there that you guys poured your drinks I've got my mexican whiskey here with me probably shouldn't because I'm still coming out of a cold. But I think we're going to talk drones. Ah. 00:10.64 Paul Oh no. 00:18.89 Paul You've got whiskey for a drone discussion or I p your liver. Ah. 00:19.71 archpodnet I Do indeed it's probably a really bad Idea. It's a really bad idea. Ah so you you mentioned you brought your drone back out and we've talked about um one of the other one of the other people out there brought a drone out as Well. Two very different drones doing 2 Very different things. So Let's have let's talk about what you guys are doing with those. 00:41.30 Paul Right? So um I was mapping at a macro level. Not not fully at the level that I did it over the fall when I was mapping the entirety of the site but I would fly at at one hundred meters different blocks of the site and. 00:56.36 archpodnet Oh. 00:59.32 Paul This was for a variety of reasons. Um Chief amongst is that the soil conditions It was not rainy when we were there but it had been raining prior to us getting there meant that the so the ah the architecture just subsurface the ancient architecture. 01:08.85 archpodnet Okay. 01:17.51 Paul Was really visible from the air. So I would I programmed a couple routes that would fly blocks of the site that I would do you know every couple of days and then I would take all those make ortho mosaics of them with drone deploy and. 01:17.66 archpodnet Oh. 01:34.41 Paul Hand those over to the to the site architect who was going absolutely blind nightly not because of that going blind nightly because he would sit up. Um, looking at these ah these ortho mosaics and plotting out where he could see walls and streets. And other features of various types. You know subsurface you can't really see them when you're walking on the site in some cases you can, but for the most part you can't but as soon as you get up one hundred meters some of them pop I mean there's no denying. You're like oh yeah, there's a house There's a city wall. There's this, there's that I oh look at this street. It runs through this whole neighborhood. Um. 01:58.93 archpodnet Yeah. 02:14.60 Paul All sorts of great stuff like that and this isn't anything new. This is stuff that people have been doing with ah with drones to some extent but definitely with ah with aerial photographs like Corona Imagery for quite a while now but it was really effortless for me to just. 02:25.45 archpodnet Yeah. 02:32.89 Paul Program this flight route and do the exact same flight route today tomorrow first thing in the morning end of the day. Whatever and then just hand them to to Mark the the architect and let him work on what he could see so we're going to circle back with that. Also see if how that. Interrelates with any of the stuff that we found on the surface survey right? if we can start to see patterns in the kinds of buildings that there are in relation to the kinds of artifacts that we find in in the surface survey. So you know that that's what I was doing on one of the ah the flights i. 02:59.32 archpodnet Over. 03:11.62 Paul Noticed ah part of the city wall and we know we're part of this parts of the City wall is but this is an extension that we hadn't seen and it was so clear that my jaw dropped when I saw it on the on the screen on the ipad while I was flying. Um, so. 03:22.12 archpodnet Um, yeah. 03:26.87 Paul You know, ah the ah the reps were nearby working on their surface. They're part of my surface survey and and so when they finished up I called them over to go show them and they were really excited too. But after I saw it from the air. We could see it crystal clear on the ground I mean you could practically cut yourself on these lines. They were flat. 03:35.67 archpodnet Um, yeah. 03:41.00 archpodnet Nice. 03:46.52 Paul But one part that's where the wall is and here's where the city gate is and here's inside the wall and there's the glassy or something on the other side. So um, so that was really exciting that we could use this real time It's not like going and looking at old satellite imagery old spy photos. Whatever it was hey could you check this area. Um, and so then alongside with that same you know, kind of mid-level. Ah drone scale drone mapping zaid was also really interested in what's happening off of the main tell of Lagosh because there are a number of small features small tells all around. 04:19.27 archpodnet Yeah. 04:24.48 Paul Ah, so I flew those so that he could have those mapped as well and he identified a few new sites that had not been previously documented including 1 which was really thrilling to me ah had a fair amount of uruk period um, ah pottery on it. 04:41.77 archpodnet Okay, no. 04:43.53 Paul Which is really amazing I mean you Don if you don't know what the oru is. It's um, it is the time period of the fourth Mill Bce where writing is developed where these cities really congeal into these big things. Um, so this is very old and very. 04:53.20 archpodnet Okay. 05:02.93 Paul Interesting in mesopamian history and I keep on using the word interesting I should find a better work. Ah, very important very significant in in mesopamian history. But also to a large extent to world history beyond that because this is some of the first time that people are doing this sort of stuff anywhere in the world. So ah, so finding that site and being able to Map. It was also really really cool and we mapped it again effortlessly from the sky. Um, the other effortless thing and this blew me away more than anything I did frankly. 05:26.38 archpodnet Oh. 05:35.15 Paul But like I said Sada had brought in this dja mavic mini the the original mavic mini which she has done to other projects. She's a very active field archeologists and ah so I like what are you going to do with that said, oh I'm I'm going to map this yeah, the the trench at the end of the day and that's what she was doing. She was just. 05:37.27 archpodnet Um, yeah. 05:54.87 Paul Popping it up in the sky flying it for 10 minutes and taking a couple dozen shots of her trench from a bunch of different angles bring that back down into the lab at the end of the day load it up in in meta shape and generate photogrammetry of her trench day by day which. 06:03.90 archpodnet That's awesome. 06:10.15 archpodnet Um, yeah. 06:14.26 Paul You know I know that people do that but I didn't realize the kind of casualness that she could bring to it of just yeah, this is what I do you know it's It's no different than you doing a sketch map at every day except for it's much more highly accurate than that sketch map and you can rotate it around this way and that way. 06:20.40 archpodnet Um, yeah. 06:31.40 Paul And if you're doing it every day you can compare today against tomorrowgainst the day after with colors and textures and such and then you can also hand it back to the architect who can then go and do your nice finished site plans you know trench maps at any particular phase based off of those photogometric models which was. 06:33.59 archpodnet Um, yeah. 06:51.30 Paul Wow! Real value. Add to me I I do not know why this let me see this absolutely should be in everybody's toolbox. Not sure about metashape that's expensive, but there are other options too I'm personally I'm playing with web odm at home to see if I can accomplish something similar. 07:06.70 archpodnet Yeah. 07:10.93 Paul But you know again, it is the casualness of yeah this is the the tool I use and she's adept but not a tech heavyavy person you know, just it's like yeah, why wouldn't I use it I use a camera this camera happens to be mounted on an airplane and it takes a bunch of pictures I can run through the software and turn it into a model. 07:14.88 archpodnet Well and you write sure. 07:28.80 archpodnet Yeah, and it's it's great that the the drone itself is no longer like the the technological impediment to doing something like this right? It really is the processing software that you actually do something with all those images but that that in itself is you know. 07:30.33 Paul End of story. 07:46.39 archpodnet Just through time there's going to be more Options. You know Meta shape is expensive and and a lot of people aren't going to use that unless they're you know, associated with the University or something like that or can just afford it. But like you said there are some other options out there that can be used, but it's easy enough just to. Throw I've got one of those doubt D Yeah Dj Mavic Minis It actually sits up in the cabinet above our driver seat in the in the Rv and we can pull it out pretty much at any time the battery stay pretty well charged and ah I can just pull it out I can have it up flying in probably 2 or 3 minutes if I wanted to and it's just.. It's super easy to just take out throw it up in the air and let it fly and it uses hardly any energy for that effort And in fact, if she had the standard like 3 battery package I mean depending on how much yeah, depending on how much she was using it. She probably didn't even need to charge those batteries more than once every few days if that depending on how often she was doing those images. 08:30.11 Paul Which she did. 08:39.82 Paul Yeah, so they were using it on all 3 trenches and I think that they were only dipping into the second battery. Yeah yeah, and that was awesome. 1 caveat I will say about this particular model is that one of the days it stopped working. It would not turn on. So. 08:41.32 archpodnet So it's a yeah yeah yeah, that's awesome. Yeah, they're so efficient. Yeah. 08:56.16 archpodnet The whole draw to their own just wouldn't turn on wow. 08:59.75 Paul The whole drone would not turn on. Um, so since I'd already demonstrated my technical chops by fixing the washing machine. That's a different story. Um, she's like hey Paul can you fix this and I'm like well yes or no those are the answers. But if you don't mind me tearing it apart I can see what I can do so I found a couple of videos online. 09:07.16 archpodnet Ah, yeah. 09:19.68 Paul And 1 of them somebody was saying oh yeah, it's really easy for dirt to get into the power button and then it behaves this way and what he showed was exactly what I had so I tore it apart we um, ah we grabbed a bunch of alcohol white pads and I squeezed that isopprobal alcohol because we didn't have any. 09:22.40 archpodnet Ah. 09:37.73 Paul Ah, into the the button and worked it through there and let it dry off and it worked fine and so then after that they did 2 things 1 is that they always kept it in a plastic bag when it was out on the site because it's a very dusty site and the other is that Mark started doing a. 09:48.15 archpodnet Yeah, yeah. 09:54.80 Paul You know a cleaning routine every couple of days on it blowing it out cleaning up the ah some of the moving parts with some of the electrical spray I don't know if you've seen that but people use it on things like potentiometers that they could crackly if they if they get a little corroded right? So um. 09:55.55 archpodnet Um, okay. 10:04.49 archpodnet Um, yeah. For sure. 10:13.11 Paul So he was using that and and it worked great. The whole rest of the time. So that's what I would add is the caveat is that if you're using something like that. You know be casual about it. But if you're a little too casual then you'll run into trouble. 10:17.38 archpodnet Um, yeah. 10:25.73 archpodnet Well yeah, some feedback on the last episode of this of this show on with ah Daniel I forgot to pronounce his last name carvallo I think it was ah he is doing robotics and archeology robotics as related to archeology and he's literally building a robot. 10:44.13 Paul Then. 10:44.25 archpodnet In Portugal where he's at and he's one of the feedback pieces that we got from one of our longtime ah longtime fans and somebody who's ah, who's an engineer and and knows these types of things. Um, he's like yeah your biggest problem is going to be dirt. He's like it's great and I'll getting a robot out to do archeology. But. How are you going to keep it clean because these things just like fail immediately when they get you know when they get dirty and ah, that's and when you said the the drone wouldn't turn on as ah as an avianist technician I was going to be like yeah if it doesn't power on. 11:04.88 Paul Ah. 11:19.58 archpodnet It's got 1 of two problems. It's either not getting power so it's in the power bay those contacts are dirty or something like that or it's the button that actually turns it on I mean those are really you know the 2 places to start I mean the worst case scenario is it's some circuit board failure but it's almost never that right? It's going to be something a little more a little more simple. So but yeah, keeping that stuff clean is a good. 11:24.17 Paul Yep. 11:32.58 Paul No yeah, well the trick with the the dgi stuff for anybody that it hasn't used one is that to turn on or off their their equipment their drones in particular. 11:39.30 archpodnet Ah, good lesson for everybody to listen to. 11:49.48 Paul It's like hold it for half a second release for half a second then hold it for 2 seconds so it's this whole funny timing dance that you have to do with the power button. Um, and I've used a bunch of their equipment and some of them are easy and some of them are extremely tricky and so this wasn't ah it was working one day and then stopped working. 11:52.35 archpodnet Um, yeah. 12:09.20 Paul Got a little finicky and got a little more finicky but whether or not it would be finicky. It was hard to tell because was it the first press that was not depressing fully or was it the second long press that was not depressing fully so it it didn't act like a power button issue per se. 12:10.97 archpodnet Oh. 12:20.77 archpodnet Ah, right. 12:28.64 Paul Because the power button essentially is 2 power buttons. Yeah, because of the way you have to operate it so it was it was a little tricky to diagnose. But as soon as I saw that video I was like oh that's that's it and. 12:31.19 archpodnet Red Red Yeah, that's tough tough on those? yeah. 12:44.64 archpodnet Yep. 12:45.60 Paul And that one on that video was really disgusting on the inside ours was just just mildly dusty. 12:48.94 archpodnet Yeah, yeah, nice. Well as you caught it ahead of time is what you did so and then had procedures around it. So all right? Well what? What do we have left to talk about in your procedures out there. What? what would you? What would you change. Yeah, you guys are going back out again. Um, are you did everything work. Well enough that you want to continue on with your methodology. Are you going to bring something new or or or lose something that didn't quite work. Well. 13:15.40 Paul Yeah, so in terms of the the surface survey the the data collection worked brilliantly I'm really happy with that. The data processing I had you know like I said I started building out a form in wildnode. Actually we used it but I had enough requests. Ah for different. 13:28.48 archpodnet You. 13:34.95 Paul Things on it that it got a little unwieldy I couldn't graft them in in a way that I wanted and then exporting the data meant that I had to remerge data in ah in database afterwards. None of that was a problem. Um, but what I would do is I would redesign that. 13:37.24 archpodnet Yeah. 13:52.80 archpodnet Um. 13:52.89 Paul Data processing form going forward. We just fine because I can you know finish those last twenty nine or 30 bags. Whatever it was and then move to a new form. You know, just reaggregate the the questions in a slightly different way and that'll be easier to process. Um. 14:02.16 archpodnet Sure. 14:11.11 Paul The other thing that probably was the biggest failure was we brought along a thermal imaging camera a a flir boson and we also purchased a gimbal setup so that we could attach it to that that Phantom drone. 14:19.29 archpodnet Yeah, yeah. 14:28.47 Paul And fly it and do some thermal imaging the idea being that we could see with the with the naked eye while with the camera mounted on the drone. We could see the the subsurface architecture we wanted to see if there was a if there were different heat signatures for the subsurface architecture versus the matrix that they were in and the answer is yes. But. 14:41.95 archpodnet Ah, okay, yeah. 14:47.88 Paul I Had a lot of trouble with that gimbal system I'm not going to throw the company under the bus but the instructions were terrible. Um, it was not balanced right and could not be balanced right? There was a way of like losing a screw and adjusting the camera left and right to to adjust the ah the roll and. 14:49.99 archpodnet Look. 15:04.29 archpodnet Or. 15:06.50 Paul I Could not adjust it far enough for this very light flier Boson camera. Um, so I I brought all that back and I'm going to redesign all the pieces. So last night I was making a new a new arm for the gimbal a shorter one so that the balance should be better So we'll see how that works. 15:09.90 archpodnet Her okay. 15:25.71 Paul Another issue that I ran into it is I got a um ah the lens on the the boson has a 13 degree field of view which based off of what I could find online what I could find on the website of the company that that sold the camera I thought that was an appropriate an appropriate angle what it turns out is is that it's way too narrow of a view for doing large neighborhood level kind of views of the site what it did work well at was ah was site. You know these 10 by tens. 15:45.58 archpodnet Oh. 15:55.52 archpodnet Okay. 16:02.62 archpodnet M. 16:03.97 Paul On the trenches rather um and we could actually see subsurface architecture that had you know that they were going to run into tomorrow and we could see the continuation of walls that we found in the trench outside of the trenches but I had a lot of trouble again with that. Um, with that gimbal setup. It would freak out and go all twitchy and then point who knows which direction So I'm I'm redesigning that another problem related to the the poor instructions was that I was told that I could capture the video on a video screen a little handheld video monitor So transmit via Radio. From the from the drone down to this monitor and I could capture on that and I could but the quality was really bad, very pixelated. Lots of artifacts. Lots of random lines running through it was not happy with that at all the. 16:44.19 archpodnet Oh no, yeah. 16:59.19 Paul Control either for the gimbal or else the radio for projecting that imagery back to the ah to the the monitor interfered with the controller of the ah of the drone. So if I'd get it past about one hundred meters from me the ah the drone would lose signal and then try to return home. 17:14.72 archpodnet Ah, geez. Yeah. 17:18.20 Paul Yeah, not exactly ideal ah toward the end when I'd basically given up on anything to to counteract the balance. What I end up doing is I took some very thick metal wire that I found and I made a little bracket that I glued on the gimbal that I attached to ale a battery to as a counterbalance. 17:36.23 archpodnet She's yeah. 17:37.78 Paul And that that worked okay, but obviously that's that's klugy beyond belief. So that's why I'm redesigning parts of it. Um, and yeah so I wanted to see if I could capture the video directly on board on the airplane. So I programmed a raspberry pi. To capture video off of the second video port from the camera because it has 2 and that worked well and so I came up with an idea of using raspberry prize zero to do that next time. Um, and then I started googling. Maybe there's a device I can buy that'll do that I can mount directly rather than have to do this with the raspberry by 0 and. The extra battery that that's going to have to bring and so on and so forth and I found an article about somebody describing different things and 1 of the things they had in There was a picture of the of 3 different models of this onboard dvr kind of device that you could have and 1 of those models I recognized. 18:15.33 archpodnet Yeah. 18:31.75 Paul Because there was an empty case of that in the box with the gimbal I had no idea what it was. It was just a clear plastic case. No markings but I'm like oh that's the same clear plastic case that oh and I went and I looked in one of the little buried circuit boards. 18:34.75 archpodnet Oh geez. Oh. 18:48.54 Paul On that gimbal assembly was a little onboard dvr. So you know the solution that I concocted with the Raspber root by zero was unnecessary I could record things directly on the the dvr on board. So um, so I guess that's good to know and long story short. 18:50.79 archpodnet Oh Geez What the heck. 19:01.70 archpodnet Yeah. 19:06.95 Paul This was not a success but I learned a lot and I have some ideas that might make it a success. Um, you know when we go back in the fall. So I'm redesigning parts and hopefully you know we'll have some better results. Yeah, next time. 19:19.58 archpodnet Okay, all right? Well that is just about all the time for this show I'm sure that if anything else comes around on this and we want to bring it Up. We'll talk about it in future shows. But otherwise yeah, this was really cool to have. Kind of bookend and then and then have the middle portion here like on the archaeology show I'm really looking forward to actually editing that here in the next couple of days so I can hear it because that's going to be really fun. Yeah, and ah, well you know the thing people got these things people have to remember is I mean this podcast. 19:45.76 Paul Ah, be careful what you wish for. 19:56.30 archpodnet This next this episode of the archeology show was recorded I mean in in real life right? and not not sitting in an office with ah with a microphone somewhere is recorded at the field in multiple areas and an airport and just ah, you know so to to ah. 19:59.62 Paul Um, yeah. Water buffaloes in the background all sorts of stuff. 20:12.70 archpodnet Exactly exactly so that is pretty cool and I hope people understand that so all right? Well yeah. 20:15.18 Paul Yeah I hope that they get the content of it. We had great discussions around what work we were doing and and again it was such a good team to work with. They were really enthusiastic and helpful and supportive of each other so it was it was a lot of fun to just sit back for half an hour and 20:24.93 archpodnet First yeah. 20:32.86 Paul Chat about what we thought about what we were doing. 20:33.55 archpodnet Yeah, for sure, cool all right? Well thanks a lot Paul This has been awesome and I hope we can I hope we can again when you go back out. Do it again sometime I know you got some other interesting field work coming up too and maybe we can talk about that as well. 20:48.21 Paul Um, certainly well. 20:51.26 archpodnet All right? Well thanks everybody and we will see you next time. 20:53.71 Paul Take care.