PCS Field Data Collector
CartoPac Mobile
Archer 4 Handheld Data Collector
Shop Online
Bullhorn WebTranscript
Introduction: The Digital Twin Concept
Glenn
Thank you — I want to talk to you a little bit about spatial accuracy. There have been a lot of changes in technology, but first I want to introduce the concept of a digital twin — I think it’s becoming a more common term. It’s really where companies are striving to create a digital model of their physical infrastructure.
In this example, you see the data collector down in the ditch — he’s got his survey pole, he’s collecting that data — and over on the right, you see that information loaded back into GIS. With a digital twin, if you have an accurate representation in your GIS system of your assets in the field, there’s a significant number of advantages: it’s safer, there’s less risk, more productivity. If we ask what the core parts of a digital twin are, it’s the what, the where, and — what we’re going to focus on today — who collected the data and how it was collected.
Terminology: GNSS Constellations and RTK
We can’t have a technical discussion without introducing some terminology and some alphabet soup. If people refer to things as “GPS,” that’s really old-school — GPS is just one of four satellite constellations being used. There are now four constellations up there: GPS, GLONASS, Galileo, and BeiDou. Having access to all four means we have access to a lot more satellites to collect information from.
RTK is a technique using the receiver to get one-to-two-centimeter accuracy in real time. RTK base stations are a known location that broadcasts a correction, so you can combine what you’re getting from your GNSS out in the field with a real-time correction, giving you real-time accuracy out in the field.
If you put a network of those base stations up covering your service territory, you’re able to get that real-time correction anywhere supported by that network. If base stations aren’t available, the data can still be collected and then processed back in the office, using information from those base stations after the fact.
Satellite Access and Accuracy Comparison
Here’s a graph — if we had just two constellations, we can see over time, as the satellites move around, the number of satellites available bounces from about 11 to 18. If we look at having access to all four constellations, we can see at times as many as 30, and the lowest we’d get is in the high 20s. So having access to more satellites means I can collect highly accurate data in more environments.
To explain this a bit: you’ve got the satellite up in the sky, and its signal is being received by the antenna of the field user, and that same signal is being received by the base station. The base station takes a correction and sends it via the cell tower and internet access, providing that real-time correction to the field user, who gets that highly accurate data out in the field. These are really the four things needed to get that real-time correction: your satellites, your base station, your data collector out in the field connected to the internet, and that cell connection.
What This Looks Like for a Field User
It’s a staggering amount of technology — all these constellations, these base stations — but from a field user’s perspective, it just needs to work, and it needs to communicate back to the field user that the system is working as designed.
If I look up in the upper left of this screenshot, I see a box saying 17 — that’s the number of satellites I have a signal from. I only need four to get a position, that’s the minimum. So I have 17 satellites, I’m connected to the internet, and each of these circles is like a cone looking up to the sky — the small circle is what’s straight overhead. What’s great to see here is that I’ve got a really good distribution of satellites — they’re not all on one side of the screen, they’re not low on the horizon. The ones on the outside are going to have the greatest error, because they’re coming through the most atmospheric interference. This tells us we’ve got a good distribution, and it’s color-coded by which constellation each satellite belongs to.
Probably the most important thing from a field user’s perspective, for knowing the system is working, is the estimated accuracy shown here — 0.1 meters. So right now my estimated accuracy is… one me— I mean, 1 centimeter. I’ve been doing this for about 20 years, and if you’d told me that with mapping-grade GNSS survey technology I could go out in the field and get a real-time signal of plus or minus one centimeter, I’d have thought you were crazy.
From a field user’s perspective, they can look and say: great, I’ve got green lights for everything, I’m connected to my base stations and my network, I’ve got my cell signal, my GPS is working well — now I just need to collect the data.
The Gaps and Overlaps Problem
As accurate as this is, what’s showing on the screen is that every second my position is being updated — the scale bar at the bottom is half a foot, and you can see it moving slightly each second, so it’s not the exact same position each time. That creates a problem for loading this data back into a GIS system — we call it gaps and overlaps.
No matter how accurate my GPS/GNSS position is, it’s not connected to the other assets. In a real-world scenario, if I’m going to put a tapping tee on a gas distribution main, that tapping tee has the same position as where it connects to the pipe — it’s not a few inches off, it’s not in a different direction. So we either create a gap, meaning my GPS position falls somewhere between where it is and where I’m supposed to be, or we get an overlap — and these have to be resolved. In the GIS systems of today, there’s a lot of talk around a connected model, meaning everything needs to be connected so we can look at the logic and relationships between all the different assets in the system.
Q&A: Dilution of Precision
[Moderator]
Glenn, [an attendee] asked if you could explain a little more about dilution of precision.
Glenn
PDOP is positional dilution of precision — it’s basically an indicator that gives you an idea of how accurate your signal is. The higher the number, the less accurate. It used to be we’d like to see it under four or five, but now, with real-time correction, we’re often seeing it in the one-to-two range.
Q&A: Accuracy Splash Screen Software
[Moderator]
Glenn, there’s one more question that popped up — a moment ago there was a splash screen showing an estimated accuracy of 0.1 meters. What software was that screenshot representing?
Glenn
That’s out of the Cartopac mobile application.
Q&A: Linear vs. Actual Pipe Depth
[Moderator]
Actually, got a few more questions — linear versus actual pipe depth, I’m seeing this throughout multiple projects.
Glenn
So we’re talking about the depth of a pipe — what’s the question, Christian?
[Moderator]
This last question — hold on. So, linear versus actual pipe depth, that was the question: “I’m seeing this throughout multiple projects.” That was from…
Glenn
If I understand correctly — for a position, you can have what’s called the X, Y, and Z value, and the Z is the elevation, or the depth. Generally, within GPS technologies, it’s easier to get an accurate X-Y location than it is to get the Z value. It can be done, but it requires some additional information to be loaded — that’s one of the areas where survey technology really can capture a very accurate Z value. What we’ve seen, when we’re putting solutions out there, is that they’re capturing the depth of cover, which gives an idea of how deep the pipe is, and that’s what we’re tracking.
Coincident Geometry and Snapping
I want to introduce a concept called coincident geometry — these gaps and overlaps have to be resolved, and you have two choices: resolve them out in the field, or resolve them back in the office. We’ve seen, where we’ve deployed a large number of field crews, that they start collecting data and overload their GIS department with information, and some of those relationships that are very intuitive out in the field are hard to understand back in the office.
So we’ve enabled a technology in Cartopac called Snapping, which allows the geometry of two features to be identical. If I have a riser on the end of a service line, the location of that riser and the endpoint of that service line are the same location — there are no gaps, no overlaps, and we’re fixing that out in the field. So we collect the data, and once we collect what we’d call the target feature, we go ahead and snap the new feature to that location.
Here, I’m working on a service line, and I’m going to map my riser and snap it to the end of my service line. First, I select the service line, then I select the vertex on that service line, and I click Snap — that snaps the riser onto the end of the service line. I’m then able to capture more details about it, and also keep track of who did it and what procedures they used.
That’s what Snapping does, and we’ve also got a technology called Fast Snap — as you’re moving along, it knows the last point you collected, and you can just say, snap to the last point collected. That eliminates the step of having to select the feature and the vertex, and from a field user’s perspective, we’re seeing that we’re really boosting productivity and greatly eliminating or reducing the need for backend processing to load this data into the connected GIS model.
Metadata and Data Confidence
Once this data gets uploaded back into GIS, we’re doing a great job of collecting the most accurate, most realistic representation of the digital twin — we’re eliminating back-end processes needed to eliminate those gaps. We also collect what we call metadata — for people who don’t know, metadata is commonly referred to as data about the data, but it defines the pedigree of the data. I can keep track of how accurate my data was when it was collected, and it’s stored back in the GIS, so other people who access that information know it’s a highly accurate position and can make decisions confidently based on where that’s shown in the GIS.
As an example, here’s a service line that was collected, and the meter installed, and it tells me everything I know about that metadata — I’m just showing a few examples. I think we’re collecting — Aaron, what did you say, 43 different user and GNSS metadata attributes? That gives us an idea that we’re doing a really good job of collecting the most accurate data possible, and others can look at that data throughout its entire use in the system and know how accurately it was collected.
The other thing showing that gaps and overlaps have been resolved: if you look at our scale bar here, a foot — there’s that riser on the end of the service line, and even when I zoom in, there are no gaps, no overlaps, because those two points share the exact same location.
Closing Recap
Glenn
Coming back to the beginning slide, looking at the digital twin — we’ve gone through the collection process, talked about how accurate it is, talked about the challenges of gaps and overlaps and how we see that reflected back in the GIS information. This is from Cartopac and the solutions we deploy — we’re doing a really good job of creating a very accurate digital twin of what’s installed in the field versus what’s represented back in the GIS and other enterprise systems.
I think we can open it up for some questions, if there are some out there.
Q&A: Overlaying ILI Data for Depth of Cover
[Moderator]
There was an additional question in the chat — Aaron actually answered it. It was from Deborah, she’d asked: can you overlay ILI data to confirm depth of cover? I see that Andy answered her: yes, we can bring tabular and GIS features down to the mobile app. And Aaron as well — he said Cartopac excels at bringing multiple disparate data sources together, such as ILI, OQ, or work order data. Once that data is loaded, we can use it to perform various comparisons or data entry updates based on various data sources.
Glenn
Deborah, just to answer your question very directly — overlaying ILI data to determine depth of cover will depend on the type of survey you’re overlaying it on top of, like a close interval survey, pipe-to-soil, things like that. We can follow up with you on legitimate models that show how you can confirm depth of cover, but absolutely, the software allows you to overlay it — it just depends on the different types and quality of inspection data you’re working with. We’ll follow up with you directly and show you how that looks for confirming depth of cover. Appreciate the question.
Closing Q&A: RTK Support in FDC
[Moderator]
Does anybody else have any questions for Glenn, or anything we can help answer for anyone? You’re welcome to unmute now if you’d like. We’ll make sure we have this presentation available for everyone, along with Glenn’s contact information, if you wanted to ask questions or get a better understanding of high-accuracy GPS and GNSS location data. We really appreciate the time today, and we look forward to seeing you in June.
One more question, from our friend Rob Flor: is the FDC app capable of collecting data from a GNSS receiver connected to RTK?
Currently, the answer is: we’re working on it. RTK presents a different challenge — it’s an additional layer of software doing the post-processing side as well. If I think about the Juniper Geode 3, that’s doing real-time interpolation to get you sub-meter accuracy, but we’re working to add RTK to our field data collection application. Cartopac, and Glenn — you can confirm — I think you already have RTK integration with Cartopac?
Glenn
Correct, correct, yeah — and that’s what the majority of our field users are using, they’re out there getting that spatial accuracy. I think it’s really driven by — we often ask the question, what’s accurate enough — but with the regulations as we understand them, you’re not going to get the spatial accuracy needed from a regulatory perspective without some sort of correction or post-processing. In general, without needing RTK, you can generally get around sub-meter accuracy without real problems, but when you want to get into sub-centimeter, decimeter, all of the very highly accurate modes, that’s where RTK definitely plays a major part. So we’re always making sure our software has those capabilities, and again, happy to answer questions and follow up with anyone who’d like a better understanding of what type of accuracy we can provide with our field data collection solutions.
With that, we really appreciate the time, we’ll make sure we have follow-ups, and we’ll post the presentation as well. I’ll give it back to you, Christian.
[Moderator]
Thank you again.