CIAT: 0:03

Welcome to Where it's AT, the Architectural Technology podcast from CIAT that uses technology, to talk Architectural Technology for technologists. In this episode, we're asking, "Will buying a drone or laser scanner make you a better designer?" Leading today's conversation on the pros and pitfalls of data driven design is Dan Rossiter FCIAT, who is the Built Environment Standards Lead at BSI and Vice-President Technical for CIAT. Joining him on our digital sofa is his guest, Andrew Evans, a leading authority in surveying technologies from Topcon Positioning Systems.

Dan Rossiter FCIAT: 0:41

Hi Andy, nice to see you.

Andrew Evans: 0:43

Hi Dan, lovely to be here. Thanks for having me.

Dan Rossiter FCIAT: 0:45

That's alright, we were just talking beforehand about how I don't think actually, aside from a very small kind of networking event, I don't think we've actually physically been together.

Andrew Evans: 0:53

No. And it's quite fascinating, isn't it? It's that power of digital I feel like I know you very well. And I would imagine you probably know far more about me, than you need to know, based on various Twitter feeds and other groups that we get involved with. But yeah, it's pleasure to actually, to spend, get to spend some time in real life.

Dan Rossiter FCIAT: 1:13

Well, the great irony then that, yeah, actually talking about tech has actually brought us together physically. So why don't you tell everyone a little bit about yourself?

Andrew Evans: 1:20

Okay, so in my day job, I work for a company called Topcon Positioning Systems, and I'm an Engineering Surveyor by training, so I get involved with the positioning side of things of Topcon. There's a lot going on for that company. The primary drivers of their technology is machine control and scanning technology, measurement technology, geospatial technology effectively, and then the associated software. So my day job is, I'm actually a Product Manager for one of the software products we have there called Aptics, which is an integration platform, and that's keeping me busy. And then my virtual secondment currently is the current President of the Chartered Institution of Civil Engineering Surveyors, and the institution has two strands of membership, commercial membership, Commercial Managers, rather, and Geospatial, of which I come from the geospatial side of the Group.

Dan Rossiter FCIAT: 2:08

It all sounds very busy. So well done on keeping on top of it. Now Andy, as you know, I'm a Architectural Technologist, as Chartered ATs we use technology to solve problems all the time, and through various bits that I've done through talking about Building Information Modeling, Digital Transformation, I've been exposed to a lot of this sort of tech - you know - I've been using tools like Photogrammetry, LiDAR, done some more traditional surveying bits, played with some virtual reality aspects. There are loads of technologies people use, when people think of reality capture. As this is the Where it's AT podcast, I think we'd be remiss if we didn't actually start the podcast by asking you an important question, which is, in your opinion, where is it at? What is the zeitgeist at the moment? What are the current trends, and what do you think's the really kind of sexy, exciting stuff that's happening in our sector at the moment?

Andrew Evans: 2:54

So, depending on where you come from, from your reality capture exposure, then you know, you picked up on the Photogrammetry element of it, that's probably the most accessible now. Interestingly, when I started out in my career in survey, photogrammetry was probably the least accessible tool - you know - because it relied on aerial surveys and people chartering flights to do all that stuff. Whereas now it's, it's handheld, it's drone based. There's a number of ways of capturing images, but the technology remains the same. It's effectively overlapping photos to create 3d models. And I think that's where we're seeing the industry go. You've got photogrammetry there, and you look at LiDAR and laser scanning based things. So as you well know, LiDAR light detection and ranging and laser scanning are effectively the same technology. I'd describe it as effectively Pythagoras and trigonometry, automated for you, and physically measuring something. So the computer's there doing all the maths, and it's sending out that beam and saying, I know where it went, I know how far it went, and I know what direction it went in and based on that, I'm then creating my 3d map on my known location. And that's all the laser scanner's doing. But it all falls under that reality capture piece, even just straight photos, depending on how you use them, a reality capture as well. And then we can venture off down the discussion on is it, are you capturing reality? Because are you actually getting smells and other noise and all that sort of stuff. Yes, you are in some situations. You look at monitoring projects where there are sound devices out on site, and you do need to know where they are, and they are picking up, you know, listening, for noises on of construction going on, or whatever it may be. But, yeah, where's it at? Convenience. So you're seeing tools that are easy to use, more accessible as effectively a democratization of tech for positioning purposes. Everything happens somewhere, and enabling that location to be used to inform us of information that we need in our day to day lives is what's changing. What does that mean things like iPhone apps, or mobile apps that are enabling you to capture representations of your surroundings digitally. You'll be aware of what that brings to architectural practices, working practices, and I think that's where you'll probably challenge me, is on this stuff's easy to use, does that mean it's any good? And that's what I'm trying to defend every day. But I'm really pleased to see that this stuff is getting into the hands of everybody. You know, you can pick up you pick up your phone, you can create 3d models of whatever you like. With a caveat, of course.

Dan Rossiter FCIAT: 5:35

I personally found it interesting, because when I started design and practice - you know - I did traditional surveying techniques. We had an in house Surveyor, I worked for my Local Authority, and the 'always rule' used to be, is that if you asked him to do a job, you had to go out and help him, because he was the one man who could do it, and no one else could. I've gone from very basic surveying to watching all this technology slowly come in, and it's incredible. The fact that you can go around with your iPhone and start to do reality capture is fantastic, and it is great to see how it is coming to the masses. But that's where the danger is. There's so much shiny, shiny out there, and we we're all weak to it - we're magpies - we look at something cool and shiny, and we want some of it. Once you have something, you want to play with it. And this is the problem with things like drones and going out and purchasing some of the software, is that people will buy it and think, right, well, actually, I can now have a scan of everything. I don't need to send surveyors out to do stuff, I can just fly this machine over it'll capture everything but actually, people have then had - just because they can doesn't mean that they should. Technology seems to be thought of first, and people will go, "Ooo I'll buy one of those." Or "Ooo I can get this now", or, "Ooo I can get - you know - a full color, high fidelity super scan of my house". I'll go, "Okay, cool. What are you going to use it for?"

Andrew Evans: 6:49

And that's a valid question. I think there's a number of things to unpick, from what you're saying there. The catch-all response to that is"Ask a Surveyor", and you'll get the results you need. But I'm acutely aware of - there's issues - I was chatting with somebody earlier this week, actually talking about surveys that they had, which were combinations of aerial LiDAR with road based data capture, with other survey data, and getting all that data to stitch together - it doesn't happen. You know, the coordinate systems need to be understood, what the potential errors in that data can be. So - you know - in terms of, say, a building scan, you're going to get reflections, you're going to get stray points, you're going to get all this stuff that needs to be understood and corrected, or at least categorized as uncertainties in the data set. You know, you hear horror stories of people procuring point clouds, receiving these point clouds, and then - you know - what's interesting with the CICES Members is - and Geospatialists in general - very quick to adopt that shiny technology, probably distracted by that technology a bit and going off and doing amazing stuff with that. You know, there's a lot of exploratory work getting done. So there's some fascinating tech being used for mine surveying, for really challenging environments, where, where this technology lends itself well, so this remote sensing side of things. We used to talk about sort of extreme survey - you know, if you want to sort of big it up a bit and if you liken that to base jumping - how are you going to measure a building, an antenna, a span, or the Earth. You know, that's something I played with for a little while trying to imagine and describe why you might want to scan something. It's difficult to access. It's difficult to capture because it's a wonky shape, or a little bit sort of random in shape. It's not it's not consistent. And that's where the scanning technology, for example, or photogrammetry, helps capture those difficult, awkward subjects, well. But yeah, as you hinted, it's that element of- you know - how do you ensure that the data you've got is actually fit for purpose and precise, accurate - you know - we can spend hours talking about this, couldn't we?

Dan Rossiter FCIAT: 9:00

We could and what's coming across is that getting a laser scaner isn't going to make me a better designer. But I take the point that, while you know it might not help a designer do design work, actually, what we're saying is also it takes a village, and in the same way that your members aren't going to try and do the technology of architecture and try to understand the building, likewise, perhaps, if you're trying to get those hard to reach places, or try to get the kind of accurate position in what's happening, perhaps it's our role to engage with the Surveyors, to ensure that what we're using is right and

Andrew Evans: 9:32

Yeah, yeah and I think that's right, picking up appropriate. on your point there you know - it takes a village. You described earlier, helping out your your survey expert when you were learning; go and holding the pole, being the chain man. Some of the early work I did in survey and Topo survey, one of the projects we had was just a tree survey, and literally, where are the trees, and what trees are they? And the problem with that is that I haven't got a clue I'm not a tree expert. So you end up having to pull in an Arborist and to help you, just for the identification, it's in the way that all the tech that's available are just tools in the toolbox - that you need to understand how to use for the right purpose - there also are skilled people that should be engaged to ensure that you get the results you need. So does it make you a better designer? I don't think it has any impact at all on your design capabilities. What it does do, is it helps you make better decisions in your designs.

Dan Rossiter FCIAT: 10:22

I guess it'd be quite interesting to hear a bit about - you know, - how heavy and chunky can some of these files get, and at the same time - you know - how do we prevent that from happening, to make sure that you can actually open the files you might have gotten?

Andrew Evans: 10:34

It's a classic problem. So when I first got into scanning, which was around 2004 the exposure I got with Cadw in the National Trust on their heritage projects. The sad thing was, we had these amazing 3d models in point clouds that were sat in a CAD environment, with specialist software for handling the point clouds.Even then, you could only load up bits at a time. So you had to sort of select the area you're going to work on, then it would load in those points. So already these point cloud files were - even back then which was a 3000 or 4000 point per second scanner then, and so, like not a lot of data, that was still, you know, pushing gigabytes of file sizes, which back in 2004 was where you know that was not easy to work with - immediately you were dumbing down the data just to get something out of it. Which also meant that it was totally inaccessible to anybody that would didn't have that software available. But you did see viewers coming online, and there was, there were companies that were specializing in in maximizing the compression of that data so that the computer systems can work with it, and and you've seen that step with with the way gaming technology has improved, that actually, I think the data bloat is probably something that historically was a problem. I think it's becoming less of a problem now.

Dan Rossiter FCIAT: 11:56

Some of the stuff that I do outside of, outside of the job, and everything else is sometimes I look at entries into some awards, and I remember seeing a fascinating one about an historic bridge. And during lockdown there was a need to do some of the surveying out there, and actually using some of this technology, because you don't want to hang a Surveyor underneath a bridge to be able to do the measurement, particularly when they are historic and you are assessing structural integrity or those sorts of things. You know, quickly those risk flags are starting to fly up. And actually, when you start to think of why you need the survey in the first place, exactly as you talked about in that way, of, is it tricky to get to - you know - how what is the, what is the right mode of access? And you know, what are the risk factors involved? Because you don't really necessarily want to hang all these very clever people off the edges of bridges when you could maybe get a fly through, at least get an understanding of it, to capture some of those aspects. But no I think you're right, it's the input. And if you think of design as a process and - you know - stuff goes into it, and a beautiful, well thought out, scientifically valid, energy efficient, sustainable design comes out at the back of it. Things that go into that design are things like positioning, survey info, historic information about the building or the site, information about the stakeholders, intended uses of stuff. So actually, it might not make me a better designer, but I can see how it can make the design better, having better information about the surroundings through a survey.

Andrew Evans: 13:18

Yeah.

Dan Rossiter FCIAT: 13:19

Obviously a lot of our members, wthey do do design, they do other things too, and you can expect Project Management being one of those. And actually, as you're talking, I thought it was very interesting in that way of the use case of that. So when you're doing those hybrid models, combining - you know - your laser pointing with your photogrammetry in that way, where I assume it's almost you're overlaying the photo on the points to try and get that sharpness -

Andrew Evans: 13:39

A bit of both, to be honest with you, yeah.

Dan Rossiter FCIAT: 13:41

But then it means - as you say - you can use it as a record of evidence. So actually, what you can then do is you can almost go, well, here we were at three months, six months, nine months sort of thing. You could probably use them in that's where you get those kind of igloo experiences, maybe where you can actually have the client walk through it, and you get those fantastic - because I've seen some contractors do it where you have almost like a photo, and you can slide back and forth over the top of it. You can then focus in on a key junction or something, and you can say, this what it was, and then this is where we are now. Early BIM days, maybe a decade or so ago, there was a lot of talk about things like automated payment as well, and actually, if you could do those scans and capture how much dry walling has been done, actually, there's other use cases you can get into there where potentially you're looking at automated payment or scheduled payment aspects, and you can tie it into construction sequence modeling, where you might have taken the design model and said "Well, this is when we think things will get buil",, and then actually overlay the progress surveys on top. That was incredible from a project management, construction management point

Andrew Evans: 14:41

And it's amazing stuff, and I think what's important there is that you've hit on what's the point? Why would you do this? And where is it cost effective? So - you know- your slider, your image sliders "Do they need location?" They, you need to know where they are and where they're oriented to. But do they need depth? And that's where your Photogrammetry comes in, it gives you the depth of the photos by using overlapping images to enable that 3d reconstruction. And when I said a bit of both, what I meant by that, is that you can generate a point cloud from overlapping photos, and that's image analysis to achieve that based on the geometry of knowing where the two photos are-

Dan Rossiter FCIAT: 15:17

Okay, so I assume, then, if you know where Camera A and Camera B was-

Andrew Evans: 15:21

Yeah-

Dan Rossiter FCIAT: 15:21

You can kind of work out where the commonality is. And then, okay, I see, yes-

Andrew Evans: 15:25

yeah, and then, obviously, with more photos than just two. So that's like, real basic photogrammetry - is two photos, and you can do that overlapping - but to do that, you need a lot more information in terms of ground control- to enable that. Structure from motion is the, is

Dan Rossiter FCIAT: 15:34

yeah- the term that you would use to generate those -that information from multiple photos, overlapping photos - and then the algorithms go off and do all those computations that give you your inner orientation and absolute orientation and relative orientation, to enable you to generate depth maps from the photos, effectively, to create these point clouds from the photos. When you overlap that with a LiDAR or the scan data as well, then you've got like a confirmation, but also that overlap enables the bits that the photos can't see necessarily, but also get some of the shape where or surfaces that are difficult to for photogrammetry to recognize automatically. So shiny surfaces, for example, are always a challenge, but some of the modern LiDAR scanners are picking up the shiny surfaces better than the photogrammetry can, but they're still getting missing points. So you can get some better geometry to then map the photo better onto that resulting 3d model, but then picking on the other part you're going to which is, what is the point you can get these beautiful 3d models - and I started out in scanning, doing heritage surveys- yeah-

Andrew Evans: 16:38

working with a with a guy in north Wales, doing stuff for Cadw and for the National Trust, where they just wanted a digital record of what was there. And you see that , a classic use case for scanning, and it's wonderful to be involved with those projects, but super, super detailed, you end up with the situation that"What can you do with this data?" It's massive. How do you handle it? How is it convenient to use? But then thinking in terms of construction, which is,"How do we get the right information to the planning team to make better decisions? and"Can you do it as a repeated survey?" - and that's something that as a side project Construction Verification Initiative is a group of like minded industry guys and girls and talking about, How does construction progress work with modern methods of measurement? And that may be just a series of photos that

Dan Rossiter FCIAT: 17:26

Okay- give you the site conditions. It may be a series of photos that have geometry. It may be full on highly detailed engineering, engineering survey level LiDAR data that is really saying, you know, we had a steel frame that was supposed to be here. We've gone in. We've measured it, it's been set out against survey control. It's been scanned against the survey control, and it's not quite in the right place. But is it intolerance? Okay? It's intolerance. Sign off on that. Move on. One of the interesting things we had a few years ago was we had a new bit of kit, that we knew would work for this, and we were on a site where the guy said "Okay, you've, you've got this super duper bit of kit here. What are you going to do with it?" I said "Oh we can do -" I explained this construction verification example, and he's like "Right? I've got a problem on site in the in the corner, sheet piling. Sheet piling has gone in, and the team that was supposed to work with that following on, are claiming that it's not in the right place, has not been installed properly." Okay?

Andrew Evans: 18:22

"how quickly can you tell me and what's going

Dan Rossiter FCIAT: 18:23

Yep-

Andrew Evans: 18:23

Project Manager comes over and he says, "Oh, on?" knowing full well that traditionally the Surveyor would go out, make a few measurements, have a list of deltas, and then go "Here you go. It's, it's this good." Project Manager's gonna go, "Marvelous. Is it any good?" And then it's still on the Surveyor to say "Yes, it's all right, I've done the survey work." Whereas what we had with this scanning capability was okay, here's your 3D model of the design and design intent. There's the sheet pile in, go out and do the scan - and in this particular instance - it was 15 minutes later, we were what's this showing me?" And we said, "Well, you see that there? back in the office processing the data. And the analysis immediately showed the point cloud, the model in a visual environment- That's where it is, and that's where it's supposed to be. You happy with that?" He's like,"Yeah, okay, fine done."

Dan Rossiter FCIAT: 19:16

Playing that back in a way - or at least the imporant bit for me - what we're saying is that in traditional methods, it was still down to the Surveyor to interpret but it was onto them to say, "Is this good enough or not?" Where they might not have the construction expertise to know whether it's within tolerance or the impact it would have on the rest of the job. Whereas what we're now saying is that, because that data can be transformed into a kind of visual form, it actually allows the person who's making the decision making - in this instance, the contractor on site- is able to actually look at it and say "Well, what am I looking at?" "Well, it's here" and then he can actually make that interpretation himself.

Andrew Evans: 19:50

So, you know, traditionally, the Surveyor would go out and make a few points of measurement, like, so for example, in this room here, it'd be like, the four corners and maybe some spots on the wall-

Dan Rossiter FCIAT: 19:58

Yeah, sampling more than anything

Andrew Evans: 19:59

and sampling, yeah, and what I've just described, is the Project Managers getting to see all of those points in a visual environment, and then the else-

Dan Rossiter FCIAT: 20:05

okay- associated software can do an analysis and say, actually this

Andrew Evans: 20:09

and saying, these are within tolerance. These are, is design versus point cloud, here's the tolerance based on what you've set - so say you say, okay it will have to be within 25mm. Or 10mm. Or whatever you set. Then it's color coding the elements in the model- you might want to go and check them out. And these aren't even here. There's a number of things that a) it's giving you quality. It's also giving you progress, because it's like, well I was expecting this beam to be in and it's not there. Or even completely missing in some instances we found, in the early days. Yeah, sample is the key. Instead of it being a few points, it's pretty much everything you can see.

Dan Rossiter FCIAT: 20:48

So, Andy, as you know, our Professional Institute, a lot of our members do design work that's technology focused. So if they are looking at some sort of surveys or reality captures, what are the sort of things you might recommend?

Andrew Evans: 21:00

More often than not, I would imagine that you're wanting to spec something that's reliably consistent, clean data-

Dan Rossiter FCIAT: 21:07

yes-

Andrew Evans: 21:08

and more often than not, at the moment, that's mostly for the most robust, cleanest data - and 'cleans it' that is a technical term - would be terrestrial scanners. So, so if you're looking for internal surveys,

Dan Rossiter FCIAT: 21:18

Okay. that kind of stuff, terrestrial scanning. For the more facade based type stuff, it then comes down to, again, the clarity of the data you need, whether it's an overview - you know - you might go with a drone or aerial based capture of scan data or photogrammetry. Yes. So let's say facade based. It might be. So I can develop up some conceptual visuals for planning those sort of things. Pull in some context that sort of thing. So aerial sounds pretty appropriate there.

Andrew Evans: 21:50

There's a range of scanners with different ranges of data capture, and you might find that some of the scanners that are really well suited for internal surveys aren't going to be great for your bigger facades, though you don't have a lot of standoff-

Dan Rossiter FCIAT: 22:02

yeah-

Andrew Evans: 22:02

so getting the coverage higher up starts become tricky. Now that is easily resolved by our ever practical surveyors who just go to the building over the road, get permission to scan from the other side.

Dan Rossiter FCIAT: 22:11

Yeah.

Andrew Evans: 22:12

The alternative to that might be, then depending on the use case, if it is just a visualization analysis, then photogrammetry and maybe drone based with the right permissionss - of course. Because you just can't fly a drone anywhere.

Dan Rossiter FCIAT: 22:23

No, of course- Yep-

Andrew Evans: 22:24

you know, that's the most important thing to take then maybe a mobile mapping system is enough. away from that. You know, I'm never condoning the fact you just go and grab a drone and do what you want. There's a lot of information about it on the CAA website, for example - the Civil Aviation Authority - go and find out before you fly, particularly if you're using it for commercial purposes. We've touched on sort of the extreme structures that scanning might be better suited to; larger structures that aren't easy to reach, you might find you're going to get enough information from scanning, bearing in mind its line of sight for terrestrial scanners. What we're also seeing a lot of now is, rapid data capture and mobile data capture, which may be backpack mounted or handheld - so it's still approaching, certainly mapping and almost engineering survey grade scanning capabilities from handheld mapping. And then depending on how that data processing then depends on the But again, yeah, more info. You know where to come to get it- quality of control, that then you can achieve data that is useful, but then sort of circulate, circling back to what you were saying, it's then internal surveys, you're

Dan Rossiter FCIAT: 23:21

well of course, and - you know - if I probably going to find that a terrestrial scan right now is the way to go, depending on what you're trying to do. If it's just volume or area, room, area, gave you 1. some build up to have a proper answer or 2. Actually, we engaged the real scenario, it would be better to make these recommendations, but certainly, I think that's very helpful at a high level for people to think of the sort of tools and technologies they might use to solve their problems. So thank you, Andy.

CIAT: 23:54

One of the best things about technology is how it gets more accessible and is used in ever more creative ways. Affordable technology like this drone or the LiDAR, now built into many phones, brings previously high end services into reach and are having an impact on our industry and beyond. You're listening to Where it's AT, the Architectural Technology podcast from CIAT.

Dan Rossiter FCIAT: 24:20

Of course you don't just use laser scans in construction, and I recall fairly recently, there was a news article - and we might be able to put a link in the show notes - about something recently with the Titanic. It sounds to me like it's not just in the Built Environment is where this sort of technology gets used.

Andrew Evans: 24:34

The photogrammetric model of the Titanic, that's an amazing feat of technology to capture that using photogrammetric techniques under the sea, underwater, absolutely phenomenal. Interesting for me was the little bit of noise around that of people complaining that the company that had done the work were claiming it was a digital twin. It's a representation of reality, for want of a better description and throw some other buzzwords in there. I talk about construction every day - it's my it's my passion - but when I first got into scanning, it was heritage focus-

Dan Rossiter FCIAT: 25:10

yep-

Andrew Evans: 25:10

it enables huge communication but what I find fascinating with it is there was an amazing bit of data I'd seen from - it was like a mobile scan of a city - really detailed. It looked amazing. Looked really good. It really sort of, it described the city visually, or the other part of the city visually, showed it to my daughter, said "What do you think of that?" She went "What do you mean?" It's like, course, of course you can do that, and there's this expectation of - you know - it's like do you realize how much effort and technology has gone into creating that 3d representation-

Dan Rossiter FCIAT: 25:44

yeah-

Andrew Evans: 25:44

because you link it then to the gaming culture and the expectation that these 3d models, of course, they look realistic. You see it in the films. You know, everyone thinks that we can do the Minority Report style interfaces and all that- which you can, of course- but it's not like it is - you know - in the films. But what I find fascinating is where it's seeping into common thought and technology - you know - even this week, the release of the new AR visualization-

Dan Rossiter FCIAT: 26:10

or the Apple one-

Andrew Evans: 26:11

the Apple vision Pro, for example- you know - there's some LiDAR tech in there, which

Dan Rossiter FCIAT: 26:13

yes- is, it's just there, and it's and it's going to be used for, the way that they can communicate and be present in VR rooms. I think where it sort of goes outside of the construction is used for communication of situations. I know some members of the team, they could go public finallly, with the 3d model that was used for Westminster Abbey in the papers- okay-

Andrew Evans: 26:39

when they were showing stuff for the Queen's funeral and for the Coronation as well. The 3d model was derived from point cloud scans of course-

Dan Rossiter FCIAT: 26:45

oh, that's interesting-

Andrew Evans: 26:46

There's some chaps in the UK that were were front and center of that. Some of the projection stuff you see now. So again, I think this was going to be, it's not all for the Royals, but the projections that were done on Buckingham Palace, and the projections you see on other things, where they're making these 3d displays and projecting on the side of buildings,-

Dan Rossiter FCIAT: 27:03

oh, yes-

Andrew Evans: 27:04

a lot of the foundation work and the understanding of the situation is done from point clouds.

Dan Rossiter FCIAT: 27:09

Of course, because obviously, if there's windows or cornices and details, other bits and pieces, it's factored those in that makes a lot of sense.

Andrew Evans: 27:16

Yeah, yeah. For me, I think it's sort of just seeping in. It's like expected to communicate like that. How's that data created? And then you think about where - you know - the next step, I think, which is going to be interesting for me, is where point clouds are used as a way of navigation. So if you think about persistent graffiti in Snapchat, for example-

Dan Rossiter FCIAT: 27:37

okay-

Andrew Evans: 27:38

there's something that - you know - you go and spray a bit of virtual graffiti on the wall, and then some other user comes along, and you're positioning them based on all sorts of positioning sensor technology that's there. But part of that would be, or could be, you making use of the LiDAR that's in the mobile device there to enable you to get a relative position, so that you've got that persistency of being able to see someone else's creation in place as it was created.

Dan Rossiter FCIAT: 28:01

Now I'm not young or cool enough to use Snapchat-

Andrew Evans: 28:04

No me either-

Dan Rossiter FCIAT: 28:04

but what I'm hearing is almost virtual

Andrew Evans: 28:06

yes, yeah, you say someone. So a long time, 2010- geocaching in that sort of way, where what you're effectively then doing is that you could walk up to - let's say - the shard or something like that, and you could you could turn on the left hand side of the door and see graffiti that someone else has done through some sort of augmented reality vision - or something - you know, , that's pretty incredible. So we've talked about entertainment usage, where we heard about you, where you get those kind of overlays on buildings, you can get the fantastic presentations. It can get used in kind of the film and media industry, quite a lot. I bet Formula One - I imagine - is probably a great situation, one where someone we, I had access to a mobile mapping system, and

Dan Rossiter FCIAT: 28:42

Okay- might have gone around to to map, actually the tracks. And- effectively we had to drive around Silverstone- Oh, amazing-

Andrew Evans: 29:00

And the resulting model was used for MotoGP. Now, the complaint from the game players that were using MotoGP was that the game that they had before we did the scan wasn't realistic. So where the track was going down in real life, that wasn't represented in the model that the games had built into the to the system. So we had to - I did the slowest three laps of Silverstone ever - probably - and it was in my own car as well. I mean, this is like Heath Robinson days of mobile mapping, where the systems on my own car, I've had to set it up so it's got a leisure battery inside of it, so we can run the power mat-

Dan Rossiter FCIAT: 29:39

yeah-

Andrew Evans: 29:39

all that sort of stuff. Three laps of Silverstone at no more than 20 miles an hour. And three laps were because at the time, we wanted to ensure that we got full coverage of the width of the whole track-

Dan Rossiter FCIAT: 29:50

oh okay, yeah-

Andrew Evans: 29:51

so we could go down the middle and you'd miss the side. So we drove one side middle and then the other side, and that was enough to get the coverage there. The outcome was a really quite a compelling model. But more importantly, it had the geometry and the ups and downs and the actual layout in real life, of the model that then the game was then used as the basis for what they then derived their model from, for the game.

Dan Rossiter FCIAT: 30:15

That's incredible. So not only was it making computer games more true to life, but also then, I presume it then it could also, then data that could have then been used for the driver simulations to make the drivers better, and those sorts of things, as well. It sounds pretty fantastic. You started talking about things like virtual reality, augmented reality, and my favorite one, and it's my turn to throw a buzzword in now, which is assisted reality-

Andrew Evans: 30:37

yep-

Dan Rossiter FCIAT: 30:38

And you know, I've seen great little videos where you've had bricklayers or something, where it could almost have the building virtually represented through a hollow lens, or something like that. It's actually showing you, when you're doing a complex brick pattern, where the brick goes next. So it's not just that you see the real world that is overlaid, but it's actually showing you what to do next. It brings in the work process. Hence that element of assisted reality.

Andrew Evans: 31:01

Yep-

Dan Rossiter FCIAT: 31:01

And I think that sort of application sounds pretty cool as well.

Andrew Evans: 31:04

So assisted reality is an interesting one. I think obviously there's a lot of buzz. There's gonna be a hell of a lot of buzz about AR now, which is good thing. One of the things I have realized over the years is actually it's this assisted piece that's really important. So it's the right

Dan Rossiter FCIAT: 31:09

What we the Government Andy? Or- information, at the right time, in the right place, but it may not be even visual in that respect. However, what's underlaying getting that information in the right time in the right place is that we know where you are. Sorry, 'we' that sounds very omnipresent, doesn't it?

Andrew Evans: 31:17

No, no, no, no, the system that you're using knows where you are in relation to what the job you're doing right, and then, based on that - you know - we (there's that 'we' again) know what that process should be to install those bricks correctly. Or make suggestions on an installation. At one level, you're already seeing instances where the right PDFs are given to the right person at the right time. Or you've got remote workers dialing in, using Teams to, you know, having a remote expert that isn't necessarily on site, but is dialed in at the right time to help out. You're seeing that already in play now, in various industrial situations. But then that assisted reality piece, it doesn't necessarily have to be visual, it can be audible. It can be, you know, it can be haptic. There's all sorts of things that, but the basis of it is that location piece.

Dan Rossiter FCIAT: 32:24

as you walk around different parts of the building, the audio kicks in. And actually, as you're saying then, you're getting an audio cue as you're reaching a certain position- and then it's conveying information to

Andrew Evans: 32:44

yeah- you. You can imagine the power of it if you were visually impaired, perhaps. And they could tell you where the toilet was on the left-hand side. Or warn you that the stairs are about to show up in front of you. That actually you know you that positioning and then the cues you can associate to a position could be incredibly helpful. There's some great stuff there going on. It comes back to then, what level of position information do you need? So, you know, is, is a Bluetooth beacon enough in the- the Reichstag situation there? I would imagine

Dan Rossiter FCIAT: 33:12

yeah- it was Bluetooth beacons or something like that, that was getting you the right information at the right time. Or - you know - even, even more, sort of, I won't say crude, because that's a bit dismissive, but effectively - you know - you turn up as a barcode, and that's how you're doing your navigation around the museum, or whatever it may be - you know - all of that stuff is in play now, already. So then I'm always sort of scratching my head about how that can be applied well, for survey applications in the end, and that sort of that high precision positioning. Two decades ago, I think space was the problem. And you could sit there actually say it's not just the cost of the survey or the cost of the equipment, it's actually the cost, then of the data storage in that way of holding it. But now that we've got - you know - cloud storage and more kind of a decentralized kind of internet network and super fast broadband for stuff. You don't actually physically have to maintain it on a server somewhere in order to be able to use it. So I think it's, it's certainly starting to make it a bit easier.

Andrew Evans: 34:11

Yeah, so thinking then back to those early days again. You know, we've gone back 20 years. The standard claim was that one day's data capture was 10 days processing.

Dan Rossiter FCIAT: 34:19

Ah, interesting-

Andrew Evans: 34:20

and so that was the that was the biggest blocker for adoption of scanning, was that yeah ok it's great. It's amazing tech we'll spend a day on site, but you're not going to get anything from us for 10 days. Statements like that do have some, some element of truth to them. So it was a nightmare to actually get stuff out of it. But what we're seeing now is - you know - again, I was mentioning this mobile app that's doing it - you know - in almost real time - you know - a few minutes processing to get something back to you that you can use, and the automated registration and stuff as well. So the advent of SLAM. So back in 2004 nobody was doing SLAM(Simultaneous location and mapping) that technology has made its way into terrestrial scanners, the tripod mounted scanners. Partly driven by the mobile mapping adoption, but then sort of just taking that and saying "Okay, there's a technology, a process in there that's available for mobile." How can we apply that to terrestrial? How does that work?" And overcoming the issues technical, technical issues of making that work has been dealt with. But it means then that that 1 to 10 processing time is it's almost 1 to 1. What are your thoughts on where we're going with this? It's that sort of remote capability to measure and make decisions remotely with this data, because the context of what you're working with is so clear - in reality capture data - is there a risk for an architectural technologist, that actually not being on the ground is going to end up with bad decisions, because you just haven't got the full context, because it's still digital, it's not real-

Dan Rossiter FCIAT: 35:48

it's difficult because I think there are a couple of sides to this. And personally, I - you know - I'm very fond of the work that I did as a designer. You know, I often joke when I'm driving around with the fiance, I can say - you know - see that building there I designed that. I still hold kind of passion for some of those things. I think certainly, if we get too remote and too removed, or you get to the point where you're doing design as a process instead of design as an art, and you know that there might be sacrilege for a technologist to throw the to the a word around, design is quite personal in that sort of way. But and at the same time, if you're there physically, and this is one of the reasons why I enjoyed walking around with our surveyor, is that, you know, I was there and it helps me understand the space. It's not even the through the survey data, it's through standing there and changing my perspective. So I'm not there, as the guy who's trying to decide what this extension looks like. Or how do I resolve the interface between the new usage of the building and its current form. It actually, I'm sat there thinking of, well, how do you measure or how do you reach that? And that might actually fit into my specification when I start writing things like method statements or thinking about safe working, because if he has to get a mobile lifting platform to measure it, let alone how are we giong to replace it? So I think it helps with some of those practicalities, and post the covid years and everything else, we've gotten a lot more used to remote and being distanced from things. I think it's just about being careful that we don't become so remote that we become detached from our own design work. And I don't know what the solution is yet, but I think it's it's gonna be a very interesting bit to see where the barometer moves on this in terms of how remote can you be, where you still claim ownership over that design.

Andrew Evans: 37:32

I was in a chat last week where somebody was telling me that they'd enjoyed going through the Elizabeth Line and actually walking through the stations that they've been involved for the design work with, but so they knew exactly where they were and where they were going, but it was their first time ever walking through physically. And - you know - so it kind of mirrors that, like, where's that, where's that going? I think it's going into that stage that it is possible to design like that, but I think you still there is still a risk that you might miss something.

Dan Rossiter FCIAT: 38:01

Yeah, well when we went into this, and I think we were talking about whether laser scanning, point clouds, photogrammetry, reality capture, those sort of things, can actually make someone a better designer. And what it reminds me of is the old adage we saw a lot with building information modeling, which was this idea of considering People Process and Technology, and I think what we're sort of hearing is that if you take them sort of one by one, and away people wise, I think you've mentioned a few times you've this idea of, if you are struggling, what you need to do is contact a Surveyor. And actually, perhaps it's thinking of using the right people to begin with, but also who are going to be using that information at the end. So thinking about the consumer of the Survey to ensure that what you're doing is creating just the information needed for that person - you know - from a tech point of view, yes, there is a cost in the technology perhaps if you're going to buy it. Or a cost in getting that professional service, but being able to think about what you're going to use it for, it means you can get the most kind of efficiency and effectiveness out of that investment, which means there is real value afterwards. Not just buying ten grand's worth a kit and then realizing it doesn't help you do your job on a day-to-day basis. And then, from a process point of view, it's weaving it into the way that you work. And I think we've, we've played before with terms definitions, you know, I'm very finicky when it comes to terminology., and - you know - some people will talk about this idea that digitizing is going from paper to onto the computer, but digitalizing is where you actually change the way you work to do digital processes. And like you talked about, if you can get your phone out do a quick bit reality capture, and actually that helps you inform decisions that you're making, then actually perhaps doing that scan has made you a better designer, because you're able to actually inform and make decisions afterwards.

Andrew Evans: 39:48

Brilliant, great summary, I'd just like to add

Dan Rossiter FCIAT: 39:49

Yeah.

Andrew Evans: 39:49

And what's interesting with my niche of one thing you mentioned people process technology. I fully people that I work with, in terms of the Geospatial Surveyor appreciate that from an Engineering Surveyor point of view, and here I am as a CICES member, we talked about shiny, shiny earlier, disappear into the tech, every time. You're right, you summarize there, we look at the process, and then mentality, of taking tech and applying it. You've talked with the help the likes of yourselves, looking at Standards, looking at procedures, looking at actually documenting and communicating what that process is properly, so that everybody understands that it's not ambiguous is important. And then obviously, as I said earlier, ask a Surveyor for the people. But one of the things that we haven't touched on here is - you know - or we have touched on a little bit, is using the right tool for the job. before about using a screwdriver as a key, for example-

Dan Rossiter FCIAT: 40:40

Yes-

Andrew Evans: 40:41

Now, Surveyors are guilty of taking a bit of tech and effectively doing that. Now I would agree, you really should use a key in the lock, not a screwdriver in the lock, but we all know that a screwdriver is pretty good for jacking a car, right? So you know - don't quote me on that-

Dan Rossiter FCIAT: 40:59

some of us may have heard that that's the case. I certainly don't know it Andy. I don't know about yourself-

Andrew Evans: 41:04

and I'm not speaking from experience here, but I understand that they're quite effective. Where I'm going with that argument is there's always going to be that side application of the tech, and that's what's so exciting about it. I'm trying to think of an example now of where scanning is being used for the unexpected sort of thing. And you know-

Dan Rossiter FCIAT: 41:20

well, I think one of your examples earlier about actually the scans helping to make games better games. I certainly wouldn't have thought of that as a use case for it, and it sounds like, as you say, we don't want scan to be the solution to every problem, but it sounds like that captured reality has a wealth of benefits outside of what we traditionally think about it, As long as you can align it to processes and tie it together-

Andrew Evans: 41:44

exactly.

Dan Rossiter FCIAT: 41:45

Any final thought that might help, I guess, 1. some kind of architectural technologist, sort of person who actually wants some reality capture. Or 2. someone wider in the industry who thinks some of the stuff we've talked about is pretty cool. And where should they go to find out more?

Andrew Evans: 41:59

I'll start by saying, get in touch with the Chartered Institution of Civil Engineering Surveyors if your focus is engineering survey, certainly do that. Then in terms of the geospatial elements of gathering point clouds and reality capture data within the UK, you should also look at the Survey Association at the RICS. It's quite a niche industry-

Dan Rossiter FCIAT: 42:18

okay-

Andrew Evans: 42:18

I would imagine you would get most of the people you need throughout the country from those three institutions, particularly with that sort of reality, capture, aim in mind, I would start there.

Dan Rossiter FCIAT: 42:30

No that sounds fantastic. Thank you very much Andy, it's been a pleasure to have you on the podcast.

Andrew Evans: 42:33

Thanks Dan. Great talking to you.

CIAT: 42:36

You've been listening to the Where it's AT podcast from CIAT. All that leaves is to thank Dan and his special guest Andrew for taking us through the world of data driven design. Don't forget to subscribe and share and until the next time The contents and views expressed by individuals in the Where it's AT podcast are their own and do not necessarily represent the views of the companies they work for or of the Host. This podcast is for informational purposes only and should not be considered as advice you.