Synopsis
Eos hosts “Accurate field-to-office data sharing with ArcGIS apps and submeter GNSS receivers” at the 2019 Ozri Conference in Melbourne, Australia. In this speech, Eos Positioning Systems CTO Jean-Yves Lauture explains how to access real-time GNSS locations on mobile devices using Esri apps such as ArcGIS Collector enables reliable field data collection and sharing.
Click play above to watch the webinar recording, or keep scrolling to read the transcript …
The History of Eos
Jean-Yves:
It’s the first time for me in Australia, and it’s a real pleasure to be among you. You have a very beautiful country with very, very nice people. I’ll take this back with me to the cold winter of Canada, where I’m coming from. That’s where Eos Positioning Systems is based, a little suburb northeast of Montreal.
Eos comes from a long history. I’m not that old, but it started back in 2001, where a technical team basically pioneered GNSS receivers by coming up with the very first sub-meter Bluetooth receiver on the market — back in the old days, when nobody knew what Bluetooth meant. In that time, to make Bluetooth work you had to purchase a specific device; the big compact flash cards. For those old enough to know what a compact flash card is, it was used to make a receiver become Bluetooth.
The reason we did that was because things were changing, especially in GPS. The US government had decided back in 2000 to remove selective availability; before that, to get your accurate position, you would have to do differential correction in post-processing to remove the scrambling of the signals that existed. Because of those, it was 100-200 meters inaccurate, and you had to do post-processing using data from a base station to correct your data afterward. People were sick and tired of doing that. At the same in the US, a space-based augmentation system came up, and we could make use of that signal.
Why do you have to do differential correction on the GPS? As an example, take your smartphone tablet; you get 5, 10, 15 meter accuracy. However, this signal has to travel from 20,000 kilometers above the earth down to your phone. In that journey, it travels through the ionosphere and the signal is disturbed. This error is based on the time of arrival, so a few microseconds will generate meters of errors in your positioning. So, you have to do a differential correction where you take data from a known location for a known base station and apply those corrections on the receiver that is doing the work in the field as a principal differential correction.
I’m explaining this to you right now, so we can talk about what Australia has setup newly this year. You will understand a little bit better towards the end.
So, Eos came with two new things from the past history.
First was to remove the GNSS component from the data controller device, because by the time you would purchase a device with Windows Mobile (at the time Pocket PC 2002, 2003, Windows Mobile 5, 6) software would come out that was no longer backwards compatible. So all-in-one GNSS receivers with computers were becoming obsolete by the time you would buy them. So, how can we go modular? Bluetooth. However, the public was not ready for that. It took quite a long time for them to adopt the fact that we could separate the two components. The GP GNSS receiver would survive, and you could change your computer and your software on the go and evolve with your receiver.
The second thing was that to be able to bring high accuracy GNSS receivers, we needed to demystify what once was true only for surveyors — getting accuracy down to just one centimeter of sub-meter. To bring it real-time to the public, so anybody could use a GPS receiver with high accuracy.
Eos Today: Real-Time High Accuracy
After all these years, we can say that the mission was accomplished. We didn’t just need to pioneer having the very first sub-meter Bluetooth receiver, but also, when Collector came out that was a complete game-changer. Users could do data collection in the field without having to handle all their data, because it goes directly on the cloud in real-time. In order to do that, we needed to have real-time differential high accuracy directly from the field to the server. You don’t have to handle your data anymore; you can watch all that your employees are doing — whether they’re doing a good job or not, whether they’re going to receive a phone call from you or not.
Collector, three years ago, needed to add what is called metadata. So, not just a GPS location, but also everything related to the quality of the position that you will be taking; what was my estimate horizontal accuracy? How many satellites was I tracking? Was I in differential GPS mode? Was I in RTK for one centimeter?
All this information is now in your database with every single point that you captured in the field and every single location. So, you can go back three years, five years from when that point was taken on the field on that date and be certain that the point was good and if that if there was any problem, it was from elsewhere.
We work closely with the Collector team because we had the very first Apple-certified RTK receivers, and they use that to develop and to add metadata to Collector. So, we have a very long history of working with them.
Our receivers are cross-platform — compatible with Apple and Android. We had to work hard to get the Apple certification. It was very rigorous but in the end, it paid off a lot. I would say 75 percent of our customers are on Apple devices, believe it or not. And twenty years ago, nobody dreamed of something like this happening. Where people would be using a smart device, phone, tablet to collect high accuracy data. Also, as I said, we work very closely with the Esri team, the Collector team, Survey 123, and QuickCapture to support high accuracy data.
What is new for Australia is that you guys have one of the best upcoming free differential correction services for sub-meter on the face of the planet. You already have it today, but it’s going to be/get better. We’ll talk a little bit more about this. We can get free differential correction to get sub-meter accuracy anywhere in Australia/New Zealand. This is being put in place by your government. We will talk more about this later.
Innovating for the Needs of Clients
We innovate for our customers by supporting various workflows that they might need in the field. One of them that we did last year was laser offset measurements.
We had, for example, a water company: one of our big customers in the U.S. called us and said, “Hey, we would like to map three valves in the middle of the street but we don’t want to block traffic. Can we map them from the sidewalk?” Or a zoo, for example, that made a center story of ArcNews this summer. A gardener decided to map all the assets, which were utilities like gas, water, electricity, everything there is in a zoo. Also, animals — the gardener didn’t want to get eaten by the lions! So, he had to map all the assets from afar.
You can do that with laser range mapping and laser offset measurements. We brought that in the field, in a workflow for Collector. Another one that we released a few weeks ago was a Locator, an on the ground locate utility, for when people need to map buried assets.
The current workflow is that one team goes, locates and puts spray paint. If the spray paint survives a rain or cutting of the grass, another team has to come afterwards to map and note data, then find out the depth below cover. But now, we’re combining all these into one workflow directly in Collector in the field.
The workflow goes; you press a button on your locator, you record a location, record all the metadata coming from the locator. Also, the depth below cover. The absolute vertical accuracy of your buried asset is computed and all these goes directly into the cloud then into your map. We have also implemented the vertical data for Australia. Jure 2020 is already supported in the receivers, and you’re able to get very accurate elevation directly in Collector or Survey 123 to quick capture anything that you want
I will pass on the microphone to hear Pierre from Hema, who is one of our customers. Hema was an early adopter of the new Australian free differential correction via satellite, the alternate Australia/New Zealand SBAS system. And I’ll get back after that for more about SBAS.
The History of Hema Maps
Pierre:
Hello everyone, that was exciting wasn’t it?
Surely you should be all awake by now again. I’ve been with Hema now for a bit over than ten years, and I remember once someone came to me and said “look!” when we started capturing data. We used a handheld GPS, and someone else in the car made a note in a logbook and then later on in the office they turned it into a database. Sounds crazy, but that’s how it worked back then.
About a decade later we did get our first differential GPS which was fairly expensive, I must say. We captured all center lines and POIs with that device using ArcPad. ArcPad did get a little bit old I must say. Some of you probably also had some experience with using it. It’s a bit dated, it’s a bit clunky, it’s really hard, or it was hard, to change the user interface, and also it was quite unstable. I remember we set up egg timers or an egg timer in the car just to remind ourselves to save the data regularly so we didn’t lose any data. ArcPad was running on a Toughbook and it was a fairly expensive device. That’s basically a really heavy, bulky, ruggedized laptop with a touchscreen. And being out in the field, we obviously needed to make some backups of our data we captured every day, which we still do, but to get the data back into the office we literally swapped tapes with our fly-in fly-out cruise. So, Hema Maps’s focus changed a little bit. We wanted to support WHO, regional Australia, outback Australia. How do we do this?
Hema Maps Today
I have to go a step back because something happened about a year ago. Hema Maps merged with another company called Adventures Group from Bauer Media, and they came with a lot of rich content and a wide range of outdoor magazines. We realized there’s a good synergy, or a benefit, of merging both their rich content and our spatial content. So how do we support our outback regional Australia? When you look at the latest census data, you realize most of the population lives at the outer, well, the coastal area of Australia. So how do we drive tourism and people into our big regional Australia to promote those areas?
We have partnerships with the land council and regional communities for data collection just to highlight their assets (local assets), tourism attractions, and also to help in protecting the secret sites. So how do we drive tourism and help those remote areas? We do that with what we can do best; capturing data and creating beautiful products.
What have we achieved? We helped the Northern Territory tourism and created a 4-wheel Drive self-guided guidebook. We also launched recently a Hema X digital platform where you can, later on, create a profile, do some booking, and plan your trip with all our data built-in. Our latest product is the vector map, which is a multi-scale map. We already successfully launched a platform called GeoWiki X for the CMCA, which is the current motorhome Club of Australia, and we also launched just recently a web platform for the region in New South Wales, New England — high country.
Upgrading to the Arrow 100
So, how did the transition go to the new Arrow 100? We talked to the people from 4D Global and they recommended we try the Arrow 100 for data capture. The beauty about the Arrow 100 is it’s a lot cheaper than any competitor out there, and there’s no subscription attached to it to get the sub-meter accuracy. You can easily connect it by a Bluetooth to any mobile device whether it’s iOS®, Android or Windows tablet — you might even want to use a serial connection if you want to. It’s really lightweight and battery-driven, so you can take it out of the car, go for a walk and capture a bush walk if you like. And again, it’s the sub-meter accuracy that you get with it.
So, what do we capture with the Arrow 100 GNSS receiver? We still capture all center lines, but we also capture point error. Why do we capture point error? This is because we get a lot more extra information, which means elevation, speed, and direction of travel, which can be used for other analysis later. They can be used, for instance, to come up with track difficulty. We still have other dedicated devices to capture PRI data, photos, and videos. And again, everything with the Arrow 100 is sub-meter accuracy.
So I’d like to get Jean-Yves on stage again talking about the SBAS.
What is SBAS?
Jean-Yves:
SBAS stands for space-based augmentation system. It’s designed to augment GPS receivers, to give it better accuracy and to correct the differential errors. SBAS started with WAS (wide-area augmentation system) in North America, the blue part that you see. It’s not confined to where the base station is — instead, there’s a network of ground monitoring stations across a certain area.
For example, we take the US; there’s a network of base stations in the US, Mexico, and Canada. Those stations compute a narrow spheric model and they are uplinked to satellites. Those satellites broadcast the corrections for us on the same frequency as GPS, so the receiver can get GPS signals and also the corrections for the GPS. With our receivers, we’re able to go down to about 40 to 60-centimeter accuracy using the WAS signal, and we do other things with it, but we’re able to really nail it down to those accuracies.
There are other SBAS systems around the world: for example EGNOS in Europe, then you had MSAS in Japan, you had GAGAN in India — all these are augmented systems for GPS. Australia is currently in a testbed.
I forgot to add, those systems are primarily made for aviation, so it has to have what they call a “safety of life” or “precision approach” of the aircraft. So, it has to be extremely reliable. Before a system is fully put into use, it is certified for aviation. It has to go through rigorous testing for about two or three years.
The Australia/New Zealand SBAS
Australia decided to venture into that path finally, and they have a network of stations, but the system is in a testbed for now, until next year. Its test building has been happening since the last year, and they have a network of stations between Geoscience Australia and New Zealand — this entire area is covered. They are constantly monitoring the system while it’s in testbed. They now have outages; of course, there will be outages of a few hours or a day but they notify the end-users. You can subscribe to the newsletter and if you start using it, you’ll notice that they avoid going to the field.
Here’s the list of stations they’re currently using, and with that, you’ll see what I was talking about, which is the Ionosphere grid. It is computed and uplinked to the satellite, which is just due north of Australia on the equator. To see the satellite, you need to be within the green areas of the map to have your sub-meter accuracy. It’s free and is usable today.
What’s going to happen with this system is that right now Lockheed Martin has a contract for the testbed; it’s going to go until the summer of next year, and then they’re going go to a tender, and after that the company that is awarded the contract will implement the real monitoring stations on the ground.
Right now, they’re using some of the core’s network stations to have the testbed for the feasibility, but they’re going to have a completely independent network of monitoring stations on the ground. Once it is certified, they’ll go to the certification process, which is going to take two to three years, but again, the signal is usable. For WAS, it was usable for years, the same thing for the other systems. The accuracy is phenomenal.
Australia’s the last one to come on board with an SBAS system, but they’re also the first ones to innovate. They will have not just GPS correctors in a signal, but also Galileo; they will have two frequencies. They call it DMf mc2 frequency mode corrections, and they will be able to broadcast correctors for GPS, and go to constellations on two frequencies, and also they will also include what they call a PPP (precise point positioning) in RTK mode. There you’ll, in theory, be able to go down to 10, 15, 20 centimeters. All these are supposed to be free of charge.
So, it’s a great system being put in place and I’m very happy that someone is innovating and twisting the hands of the other SBAS systems around the world. EGNOS in Europe will follow up on the trace of Australia. You guys are doing great; I’m very happy to see that your government is doing such a great job. Back to Pierre.
Increased Accuracy and Speed with the Arrow 100
Pierre:
Thank you.
So, what productivity gains did we get out of using the Arrow 100 now in the Esri suite? We’re using ArcGIS Collector which helps us. Basically, once we captured the data, or enough out in the field, we can send it up to ArcGIS Online once we are back on coverage. Then, someone can literally pick it up in the office using tools and transfer it into our core data sets.
We have other workflows, which allow us to create our beautiful products which are atlases and guides. We also have a range of regional and state maps, and then obviously our digital products, where we have a range of apps on iOS and Android; Navigator, and then the latest product is the Explorer map multi-scale vector format which you can easily can digest into your Esri system if you want to as a WMTS system.
Talking about automation, I touched on ArcGIS Online so obviously, that makes it a little bit more streamlined getting the data back into the office. However, one of our core data sets is roads. Hema captured already quite a lot of data over the last 25-30 years, but we also have third parties whom we merge our data with quarterly.
Talking about automation, I touched on ArcGIS Online so obviously, that makes it a little bit more streamlined getting the data back into the office. However, one of our core data sets is roads. Hema captured already quite a lot of data over the last 25-30 years, but we also have third parties whom we merge our data with quarterly.
The problem we had, or faced, over the past years is that it takes quite some time to merge those two data sets. Even though it was semi-automated, there was still a lot of manual work involved and it took one person to run through all the states, about two to three years. So now we came up with a new process using a scripting language like Python, and we aim to finish the whole merge within three to six months if not even faster.
If you’ve seen all our products, you’ll notice we do print products as well. We have other workflows already in place which allows us to create those print products — it’s either also ArcPy, JavaScript, and we use tools like data-driven pages from Esri to create those atlases and guides. And obviously, we have a lot of other data layers in our database which we need to update on a regular basis and that’s where ETL tools come in place. But even though we have those pauses and plays, we always look for other ways of improving these workflows, because technology involves a lot of innovation. I’m just thinking about machine learning for instance, which will help us in the future to make these workflows even more efficient.
Thank you very much.
You May Also Like One of These Success Stories
Customer Spotlight: High accuracy adventure with Pierre Kurth
Customer Spotlight: Pierre Kurth! Eos is honored to have a customer passionate enough to travel with us to three events to share his use of geospatial tech!
Australian Organization Surveys Remote Archeology Sites
Hear how one Australian company uses Arrow 100® for heritage management alongside ArcGIS Collector. Originally presented at Ozri 2020.
