00:00.62 Paul Hi Welcome back to the architect podcast. Ah we're discussing this great little article about a new technique of mapping and part of what makes this this article So great in my opinion is that they actually test it on a real. 00:16.79 archpodnet 1 00:17.47 Paul Project you know it's a pilot project within an ongoing archaeological investigation and the site that they're that they're focused on is called horvat duvshan in Israel. It's ah north of the sea of galilee and it's a calcolithic site so it's from the middle of the fifth millennium bce to the middle of the fourth millennium bce. Um, the site itself was identified in 1970 s and had living at excavations going on there since the 2010 s and that's all stuff I knew because I read it in the article. Ah the ah the pilot project that they do with this with this hedgehog. 00:45.65 archpodnet Ah, nice. 00:53.84 Paul And marvelous Minds. Ah with the Hedgehog methodology. This ips is ah based in the 2019 season of the excavations. Um, and as a pilot project It was done alongside their more traditional total station Gnss Surveying hardware. So they wanted to compare which is wonderful Again. That's the right way to do it. They didn't they they didn't jump in with both feet and only rely on this. Um and you know and as we discussed in the first segment they are fully aware of the strengths and and weaknesses of the more traditional systems. So It's not like they were. 01:15.12 archpodnet M. For sure. 01:33.00 Paul Going into this blind which one would expect. They wouldn't be because they're our field archeologists and just like all of us we've all experienced all these different these different tools over the years um and so they they relied on what they normally rely on but they wanted to see if they could augment it and so what they did is. Ah. 01:40.69 archpodnet Ah. 01:46.55 archpodnet Yeah. 01:50.00 Paul They they set rebar into the ground around 3 trenches and then they built custom mounts with ah pipes and three d printed parts three d printed parts I've been doing that this last week to repair some stuff on some drones including the one that we use in Lagosh and yeah so I like that. 02:03.28 archpodnet Nice. 02:07.46 Paul That that that's just ah that that just tickles me that they yeah that they're 3 d printing parts in order to to test this equipment on their project. Yeah, um so the purpose of setting those rebar points is that they want to ensure consistent placement of their base stations and. 02:14.72 archpodnet M. 02:26.96 Paul As Ips System Software is and the software that they had on this. It doesn't generally care about you know world coordinates what it cares about is where in that warehouse you are right? So can use the local grid. 02:34.14 archpodnet Red. 02:40.89 Paul Software had provisions for tying it into into utms into a world grid but that was a little weird. Maybe we'll talk about that a little more as we get down here but they wanted to be able to set that out from the get-go that they could do that So they you know they set the rebar they had these things that means that they could bring the. 02:53.78 archpodnet Ah. 02:59.40 Paul The base station's home charge them up at the end of the day bring them out in the morning and plant them back in the exact same locations that they were at before so that also shows a nice level of forethought they didn't have to remap them every time putting them up on Pos which is what they did. 03:05.82 archpodnet Nice. 03:17.14 Paul Also gave them slightly better line in sight and that was something that I thought was a little interesting. It wasn't really it was only ah, kind of in passing unless I missed it somewhere when I was reading um is that it looks like this hedgehog system some needs line of sight for those sonic waves to sasonic signals to. 03:32.61 archpodnet Ah. 03:36.70 Paul To actually be usable. Um. 03:37.31 archpodnet Well that makes sense because it's time of flight measurements they're doing So it's got to go there and back right. 03:41.92 Paul Yes, yeah, so that actually to me that that kind of limits the the utility of sound for this. Um I mean that's one of the big problems that we have with ah with like we're saying before with total station surveying is that. 03:49.77 archpodnet Yeah. 03:58.63 Paul Ah, you know some of the best total stations right now that you can get are mirrorless mirrorless is almost always useless on an archeological excavation because you can't see down into the trench so you can't see what you actually want to take that elevation on so you have to put something up on it. You have basically take a prism and you put it on the ah on the pole. 04:01.75 archpodnet M. 04:17.39 Paul And then you subtract the length of the pole from all your measurements. Um, they had to do the exact same thing with their hedgehog put it on a pole but also they mounted their base stations on poles in order to maintain that line of sight. So that's ah, that's a little bit of a caveat around this whole technology that. 04:18.82 archpodnet Right. 04:37.60 Paul I'm not sure how it will be addressed because I can also imagine that warehouses Even if you've mounted these base stations up on the ceiling. You're not going to have perfect line of sight to everything that you want to track in that warehouse So again beyond the scope of this article and. 04:48.29 archpodnet M. 04:53.91 Paul Beyond the research I did but I wonder how that's being accommodated for now. Anyhow, as we said before the the there there's a modem that that gets ah that collects this data and dumps it to a computer that processes the data and it happens real time. They're talking about the basic. 04:57.18 archpodnet Right. 05:13.44 Paul Standard the default connection for your collection rate for it is 8 hertz so 8 times a second. Ah but it can be slowed down and made more accurate and I think that you know 8 times a second is a ridiculously fast refresh rate for. 05:29.90 archpodnet Yeah, yeah, yeah. 05:31.90 Paul Our purposes. Um, and so I would as ah as a user I'd rather you know even go as low as 1 hertz and and just make sure that I'm not moving ridiculously fast when I'm mapping things. So anyhow that the modem they connected to a ruggedized windows computer that they. 05:44.72 archpodnet Ah. 05:51.64 Paul Could just kind of it was a low power one. They could just fire close the lid and leave it aside for the rest of the day and they would happily chug along processing the data the hedgehog as we mentioned before was also connected to a computer in their case. It was a surface pro tablet and that sounds like something they had on the project. 05:58.70 archpodnet Yeah. 06:10.37 Paul Didn't have any direct use at the moment so they decided to commandeer it for the purposes of this and they ran qgis then on that surface pro tablet collecting the live and mea data so that they could display it in real time you know, straight off the hedgehog so that that was really cool that they step. 06:10.95 archpodnet M. 06:23.82 archpodnet Yeah, yeah. 06:28.81 Paul They spelled out exactly what the pieces were and like I said earlier I would be curious to find out what you could replace of these computers. You know, do you need that real-time display. Can you replace that with a data logger and get you know post process display at the end of the day. 06:45.41 archpodnet Yeah, yeah. 06:47.48 Paul Is that tradeoff you know what are you doing with those data you may have to even if you're getting the real-time display. You probably still have to process it in some way to get back to the excavator so they can enter say point elevations into the notebooks. Um and all this gets to just the the experimental nature of not just the technology that they're using. But also how we're going to use it. Yeah we're pretty comfortable. They're they're basic workflows that are that are used on any given site and they usually rely on 1 person that becomes the master of that particular domain the the total station person the gis person the the Gnss person the magnetomery person. Whatever. 07:07.82 archpodnet No. 07:26.90 Paul And you got to have a good sensible workflow that gets them The data gets them to process the data and gets them to return that process data to whoever is logging it in whatever format you know, Even if it's just writing it down in the ah in a field Notebook um. 07:40.57 archpodnet No. 07:43.14 Paul It's got to go through people and so there are lots of different points that any of these things could have bottlenecks and so even though they start out talking about the bottlenecks from the traditional systems I'm looking at what they're doing here and seeing for myself that there are probably also a few bottlenecks that are just different than before That's not a deal breaker by any sense. Um. 07:57.18 archpodnet Um, yeah. 08:02.33 Paul But it does have to be accommodated for it doesn't it means it's not a panacea. 08:04.74 archpodnet Yeah, and it sounded like actually that the hedgehog was one of the bigger anomalies for how to set that up because I'm I'm trying to figure out exactly how this thing collects data right? like it's ah it seems like it's.. It's finding the position of the Hedgehog hedgehog using the ultrasonic beacons and if the hedgehog is on wheels and can run itself around then it knows where it hasn't collected data and I don't know if it bumps into walls and stuff like aroomba or whatever. But ah it can it. It sounds like the beacons are like. Where's the hedgehog and the hedgehog is reporting its location and. Not reporting but you know it's basically triangulating where that thing is well obviously they couldn't use it that way so they had to set it up a different way. Yeah. 08:41.45 Paul Yeah, so my understanding my understanding is that that hedgehog is a passive receiver and 8 times a second. It says how far am I from all the ah from all the base stations that I can see and then it. 08:58.74 archpodnet Rhett. 09:00.88 Paul Radios that to the the computer with a modem attached that sees you know takes those data quickly triangulates them sends that back to the hedgehog the hedgehog then formats it as an n me a string and sends it out the Usb Port to the to the logger. 09:18.65 archpodnet Right. 09:20.16 Paul To the surface pro with qgis in their case. Um, that's my understanding of how this works I could well be wrong because it wasn't entirely clear. But if that's the case processing The the triangulation isn't really complicated math. 09:27.34 archpodnet Run. 09:36.39 archpodnet No. 09:38.13 Paul That's something that you know your handheld Gps receiver does so you could have a very basic computer and get rid of that whole radio from the hedgehog to the computer have the computer process that have it send it back to the hedgehog to you could do that with just a small onboard computer and maybe the company's already done that um. 09:53.11 archpodnet Um, yeah, just a few years ago yeah 09:56.28 Paul But that seems like an extra step in the whole process that that is not the way I would design it that that whole radio link is weird to me. Um. 10:05.19 archpodnet Well, it seemed like too though I mean the way they had to use This is first off the hedgehog is really designed to be mobile on its own right? So the the company knows and the system knows where the measuring point is on the hedgehog and it can just run around and do what it needs to do autonomously. But obviously since they couldn't do that. They. 10:13.16 Paul Me. 10:25.14 archpodnet They tried setting it where it goes but there's no, there's no Tripod mounting point. There's no stadium Rod mounting point. There's no indication on the hedgehog where the actual measurement point is and it's not a super small device. So it's somewhere on that device and if you're off, you could be off by a few centimeters. You know the width of the device or whatever. 10:29.54 Paul Right. 10:43.66 archpodnet If you don't know where the measuring point is so they basically mounted it to a pole so they could have that line of sight to the beacons and then I think it says you can you can put in a z or a vertical offset so you know how long the pole is so that can be subtracted from all your measurements but ah, but either way. The thing is not autonomous so you still kind of have the similar problem as a Gps or ah or a um or a total station except it's just easier to run where somebody's got to like run around and and drop that hedgehog on different points now the 1 good thing about this is. If. It's taking points 8 times a second and I understand why it's doing that because the hedgehogs are actually if you look at the homepage of this company really fast little vehicles and it has to be 8 times per second when they're moving that quickly because they're constantly like going back and forth and you can just really map quickly doing that but obviously a human is not going to move that fast and. You could, but but it does mean that you could map a lot quicker by moving ah really rapidly around the site and and just just continuously moving this thing around. But then I wonder like do you have to stay flushed to the ground and and say map these what if you just pick it up and you throw it over your shoulder and you're walking into a different part of the site. Is it still recording. What do those points look like you know what? I mean but there there seems to be a lot of potential data cleanup that has to be done on this unless you can just flip it on and off real easy. 11:55.59 Paul Right. 12:00.93 Paul Yeah, now if I were designing this system um base out the same basic technologies. What I would do is I would have the hedgehog with ah you know a prism pole Mount so that you know exactly where it's It's referenced to and. 12:09.41 archpodnet Ah. Yeah. 12:20.26 Paul You know the offset of whatever the thickness of the hedgehog is but but I think the base stations are basically just also dumb transmitters and then the hedgehog as a receiver, but it's also transmitting to get that signal back to the computers doing the processing again. Like I said I would move the processing on board onto the the the hedgehog. 12:26.50 archpodnet Yeah, yeah. 12:39.63 Paul And the logging. Ah, you know, maybe you do it out through a Usb port to another external device or you just log it internally but then I would just have a button on it. You know on the hedgehog that when I want to take a point I would press the button and take the point and then release it when I'm done. 12:47.33 archpodnet Yeah. 12:57.71 archpodnet M. 12:58.26 Paul Um, maybe the 2 buttons 1 to take a point you know a single point and another one to draw a line you know so I hold down the button for a while so I can trace the outline of something for example, that's how I would go about it. Yeah, it's so pretty manual, but then. 13:03.31 archpodnet Yeah. I Don't know it's It's still a pretty manual but still a pretty manual process. Yeah. 13:16.86 Paul And what they talk about and they didn't do this because they only had the 1 hedgehog but if this is the way the system works and now we've gotten rid of this bottleneck of transmitting the um, the data back to the processing computer because the processing computer is on board each hedgehog you could put a hedgehog in each trench and it becomes basically like the ah. 13:29.43 archpodnet Naha. 13:35.98 archpodnet Um, yeah. 13:36.51 Paul Ah, tape elevation ah tape measure for every um for every digger. Oh hey I need a point elevation here plop it on the point take the elevation I'm not waiting for the total station guide to finish up. Whatever you know to walk back that kilometer from the other set of so. 13:51.28 archpodnet Yeah. 13:53.77 Paul I'm not waiting for the person that that knows how to use the the gna the Gnss Receiver I'm not a bottleneck for anybody else in the process right? with just one of these little data Recorder Loggers um in each trench and just grab the one take the point and go. 14:02.18 archpodnet It's true. 14:08.91 archpodnet That is true and that is that is really nice. Um, we had ah a similar thing just for elevations I've only used it once on a site but it was ah um, a laser level. You know they'd they'd have to every every day you know the the platform that the laser level was sitting on was left out at the site. 14:19.22 Paul E. 14:28.12 archpodnet And was just a flat platform on ah on ah on a leveled known position. Um, ah Tripod right? But they'd take the device in actually so when we get out to work in the morning they would just set it right down on there. Make sure it was still level from overnight turn it on. 14:30.71 Paul Listen. 14:42.31 archpodnet And then you just had your your rod and your measuring device on there and you could quickly measure pretty much any time you want that was super cool. Um, so you're right something like that would be really nice but you know what we really need Paul we need smart trials and shovels that know their position in space and then you could just. Ah. 14:46.37 Paul Yeah. 15:01.32 archpodnet Ah, the measuring point is at the end of it and you just you just know right? It's just it's just it's all logged. Ah I don't know how that works but I'm gonna invent it somehow. So yeah I just. 15:14.47 Paul Ah, so so um, another another issue that they had is you know we said you were talking about this. Ah this laser level set up over a knownome point and that's very common. You know every construction site nowadays has something like that you know laser level that they'll set up first thing in the morning over the known point. They. 15:18.84 archpodnet Go ahead, Go ahead. 15:27.21 archpodnet Ah. 15:33.68 Paul Measure the height of it and then anything that you need ah an elevation from you've got a a consistent point of reference for everything else. So that's another thing moving from construction to archeology and I think that there probably are a number of other tools that that would move the same direction. Ah, but. 15:39.13 archpodnet Yeah. 15:49.11 archpodnet M. 15:53.62 Paul With their with this system. They set out those base stations on known points so they could always set them up in exactly the same place with the rebar and the and the poles and the ah three d printed brackets. But and then they took ah Gps coordinates on each of those all right? So they had utm coordinates. And then they could upload that in theory into the processing software so that it could process everything not as a local grid but as worldwide coordinates hopefully this has been fixed since then but that was a problem that they had repeatedly was how to get that to actually work the way it's supposed to. 16:20.51 archpodnet Right. 16:28.97 archpodnet In sure. 16:33.17 Paul And it was unclear whether it was not working for them because or it was difficult for them to work because it was poorly documented or if it was poorly implemented or if it was buggy. It may have been all 3 hopefully that's all been fixed by now because. 16:41.75 archpodnet M. 16:48.51 Paul You know we can't be the only ones who want to do something like this with worldwide coordinates as opposed to a local grid but um, you know I don't know how much that matters to companies are focused entirely on interior spaces. 17:01.46 archpodnet Sure Yeah, might not be their target audience right now. So all right? Well I mean I think wrapping this up This is ah the takeaway for me on this is not necessarily even the new technology. It's the. 17:06.40 Paul Me. 17:19.67 archpodnet Archaeologists themselves looking at something that's totally wacky and not built for archaeology at all and saying can we make this work and then not only that but taking it into the field into Israel and trying it on a site and saying what the hell let's give it a shot. You know more people need to do that so we can. Know one way or another whether or not something is just like absolutely not going to work or maybe like they kind of concluded in this article just not ready for prime time yet like it's close. It's it's getting there. There's a lot of attributes they liked about it but it's not necessarily something that's that's ready for everybody to just say drop everything and buy this. 17:50.73 Paul Is it. 17:57.77 archpodnet Um, but we do need people to you know, buy it and try it and try some of the other manufacturers out there and again, this article is written in 2019? Well the fieldwork was 2019 for this article so they've had three more years of development presumably to. You know to do different things with this and technology like this changes rapidly in that amount of time. So I really applaud them for looking at something like this and then writing up the results in ah you know in a journal that people have access to not some obscure thing that nobody's ever seen before or right? so. 18:16.35 Paul Um, yeah. 18:32.82 archpodnet Any other final thoughts from you Paul on this. 18:32.97 Paul No I mean I just echo what you said I mean it's It's great that they tried this. You can practically see the light bulb going off over their head when they saw the ah but then they went ah you know I guess the barrier to entry was fairly low because the price is fairly low. 18:43.75 archpodnet Ah, yeah. 18:52.38 Paul And that they went ahead and they tried it and then they rode up very forthrightly for us to see what worked and what didn't you know I'm actually going to really backtrack here just for a moment not on anything I just said but on another point about why they wanted to do this and about that refresh rate and this is another not exactly a bottleneck but it's a. 19:01.55 archpodnet No. 19:11.74 Paul Potential huge advance for a technology like this that can record rapidly and that's that with the total survey total station surveying software that we wrote back in the 90 s in masco we used it for drawing right? So if you want to map in a ah wall. You could take. You know 4 points to define the 4 corners of the wall or you could take more to get a more detailed view and the software would connect dot to dot and then we could also take point elevations we would do whatever but that dot to dot is fairly limited. You know if I'm taking 4 points. Um, it's going to take me no less. 19:34.54 archpodnet Ah. 19:43.60 archpodnet Yeah. 19:48.94 Paul Then a minute and a half 2 minutes to take those 4 points with a total station and that depends on the distance it depends how how steadily I'm holding the pole it depends on a lot of things but each one of those points takes 10 seconds or more then you have to move to the next one and 19:52.47 archpodnet Right. 20:03.26 archpodnet Ah. 20:06.72 Paul Each one you know and then aim the total station and another 10 seconds or more and you know so it really adds up they were using it on this project for tracing the outlines of things and because it's recording so quickly they got much much finer detail than we could with a. 20:11.61 archpodnet Yeah. 20:18.56 archpodnet Ah. Yeah. 20:26.55 Paul Jss unit or total station. So that's for me, maybe the most exciting thing in the watch this space department is that in the future in the not too distant future. We're going to be able to trace features artifacts whatever it is in the field as we go. 20:32.12 archpodnet Brett. 20:43.64 Paul And it's going to be there recorded digitally. 20:44.28 archpodnet Yeah, and that's really cool. That's that I think is huge right? It's just the the actual application of that that needs to be enhanced a little bit but the way they were able to do that that that 8 minutes 8 8 times per second comes in handy if you're using it right. 21:00.44 Paul Yeah. 21:02.47 archpodnet So all right? Well with that take a look at the show notes for this just look down at your phone or whatever you're on and look at your computer if you're still listening to podcasts on computer I don't know does maybe they do maybe they were sitting and sitting in their office or something like that they're listening on Spotify who knows but. If you have tried something totally wacky and especially if it didn't work I want to hear about it. You know what? I mean like stuff that works is really cool, but if it like didn't work I really would like to know that too because that's the kind of stuff that doesn't often get published like this article here I don't know I mean they're calling it a success partially. But also it's kind of like some stuff didn't work and that's why I said before like we really appreciate this kind of article. So let us know Chris at http://archaeologypodcastnetwork.com you can hit us up. There's other contact info on the show notes Paul's on Twitter all the time if you're a Twitter user. Go check that out and and reply to this and. Um, he's at ah, you're at lugall on Twitter right? All right sounds good and with that I think we will put it off and Paul might squeeze in 1 more recording before we'll have to find ah a guest co-host for a couple of months while he's gallivanting in the Middle East again 22:00.61 Paul Yep. 22:14.65 archpodnet And ah and we'll but we'll figure it out. We will. We will push on so all right? Well with that. Thank you everybody Thanks for listening. We'll see you next time. 22:22.44 Paul My wife.