Using Artificial Satellite Data in High-Low Mix Design Environment

SAMI HARADA: Welcome to CES501868, "Using Artificial Satellite Data in High-Low Mix Design Environment." Hello. I would like to welcome you all, and thank you, everyone, for watching. I'm Sami Harada, a senior specialist at the satellite data distribution section in an organization named RESTEC. I'm excited for you to be here at AU 2022.

MARIKO MORIOKA: Hello. I'm Mariko Morioka. I've been working in satellite-based solutions industry for more than 10 years and spending six years at RESTEC as an international sales representative. I'm very happy to talk as a co-presenter today.

Before moving to the next topic, let me introduce who it's RESTEC. RESTEC stands for Remote Sensing Technology Center of Japan. Remote sensing is a technique for observing objects from a distance, which means without touching anything. We are a non-profit organization established in 1975 and based in Tokyo, Japan. Our mission is to contribute to the lives of the people and the community by utilizing remote sensing technologies, especially from satellites. And now I will hand it over to Sami.

SAMI HARADA: Next, I'm going to read the topic I would like to cover in this class. We have two main topics here, satellite and Autodesk product. We are going to see a recent update of earth observation satellite, things we may need to know about them, and when and how we should use satellite data. Following that, we will visit how you can use satellite data from various Autodesk products.

So to begin with, I would like to start with this question, why we would want to use satellite data from Autodesk product? As you know, the world is changing, and the change has been accelerating. Climate change, COVID-19, low intensity conflict, we have countless reasons why we cannot continue with the old way of doing things.

Many organizations, more than ever, need a way to obtain updated information to every person who make a design decision. We need to think about how individuals can get access to the data that they need, empowering everyone in your organization, no matter what kind of design decisions they are trying to make and at any scale.

And that's where satellite data comes in. We also want to make sure that it will be available in any level of design workplace. That's the reason why I'm giving this presentation. OK, then we will start with recent development of satellite that may be relevant to your work.

MARIKO MORIOKA: Space map is a very common request. People want to understand the ground conditions before you make any decision on your project plan and the design. Many of you here in this class might be familiar with this type of image in this slide.

Once a satellite is launched to space, it will keep observing until the end of its life. Therefore, you can access to a satellite data anywhere in the world. The highest resolution currently available in 30 centimeters. Higher frequency acquisition, in another word, shorter revisit time over the acquisition, is one of the recent trends in the satellite industry.

The image on this slide is from Black Sky. They are a US satellite operator based in Seattle and providing the data with rapid revisit of the imagery with very low latency with a unique orbit design, constellation of small satellites, and advanced ground segment.

This image is over Utah, taken on 7 August, and just one day before this slide was created. Such higher temporal resolution images are often used to monitor the target project progress in a remote location.

Satellites can capture not only ground surface imagery, but also greenhouse gas emissions. Images on this slide are from GHGSat, a Canadian private environmental monitoring company. They found greenhouse gas emissions released in Kazakhstan mines, 54,000 kilograms per hour. Check out the link on the slide for more details if you are interested in.

This slide is about night scenery observation. These images are satellites from Channel Electronics Inc, well-known as a consumer electronics manufacturer. The traditional night observation from satellites has been limited. For example, its resolution is up to 750 meters. With this Canon satellite launched in 2020, you can get the night series in 5.1 meters, which is more than 21,000 times.

The image on the bottom of this slide was taken river in North Pole, Alaska, when it was a full moon. As you can see, there is no need for artificial lights around the area to capture the bank of the river and of water body.

The next application is 3D map. A 3D topological map is one of the most successful satellite data applications, and there are many different solutions and products. Some are available for free, but their resolution and accuracy are limited.

Among the commercial products, I'd like to introduce AW3D. They offer worldwide coverage data set with a resolution of 2.5 meters and 5 meters. The Canary Islands, Spain on the top of the image represents its 5 meters digital elevation model, DEM. On demand basis, the digital surface model, DSM, is available up to 0.5 meters resolution, which you can see on the right image of Moscow, Russia.

You can access to the data without having a permission from the authorities. You just need to tell the data provider your target location and the data specification. So it's pretty easy to get the data. These 3D digital data, including digital timeline model, DTM, has been used in many projects anywhere on the Earth, including disaster mitigation, flood simulation, mitigating climate change, water resource management, and master planning of urban development.

I have two more slides to present in this section. In the previous slide, I was talking about a roster data. This slide is about a vector data. By utilizing the capability of satellites in terms of scalability, consistency, and project management friendly, vector digital data are optimal for various simulations, including 5G network planning, environmental research, and aviation. These vector data are also from AW3D, the same as 3D map in the previous slide.

The provider completed the vector data set of Japan with 90% population coverage, which is about 100,000 square kilometer, without funding support by the government. This example clearly demonstrates that a satellite solution can be easily scaled to a state nationwide.

So this is the last slide for this section. What now I'd like to showcase is an example of synthetic-aperature radar satellite. It's often called SAR. SAR is able to collect imagery no matter if the target area is bad weather or at night using radar. It's also able to detect very small, slow movement, such as centimeter per year, about 0.4 inch per year in the target area by using a technique called interferometric synthetic aperture, InSAR.

The analyzed data can be overlaid in Autodesk products. This image it's showing here is from InfraWorks. This insert technique has been adopted in a wide range of industries, like infrastructure, mining, oil and gas, local government, and so on. That's all for this section. Sami will be presenting the next topic.

SAMI HARADA: Thank you, Mariko-san. Next, I will talk about Satellite 101, how they work. So unlike what you see in movies and TV shows, satellites will not and cannot change their flying direction. Once they are launched, they will stay in the same orbit. And the satellite personality, which is where it will visit and how often it will visit, is decided by that orbit.

So there are three types of orbits, and this first one, sun-synchronous orbit, is the most common orbit. I am going to explain how a satellite in this orbit works. A typical satellite flies above 500 to 700 kilometers above the Earth's surface and observes the target from there. This is just like you are watching what's going on in Boston, Massachusetts, from Washington DC.

So by the way, you can decide the altitude when you [INAUDIBLE]. The lower orbit, you will get a higher resolution. The higher orbit, you will get a wider coverage, also a longer life expectation of the spacecraft.

So on this orbit, a satellite can visit almost anywhere. They fly from north to south and it passes over you around 9:00 AM to 2:00 PM. It takes about 100 minutes to wrap around the globe, and it returns the same path about every five days or so. This means if you miss a chance, you have to wait for five days for the next round. So five days is really quite long time, so what we can do to increase this frequency?

We can launch many satellites, making a concentration, but without doubt, it's very expensive. Or you can point a satellite to take a look from an angle. This is called off-nadir viewing. Off-nadir viewing is a great way to increase its visit frequency, but with cost. It decreases a resolution and accuracy as well as increased occlusion in other areas.

This figure illustrates how horizontal accuracy is compromised due to the altitude of the satellite. Even when we can precisely control the posture of the satellite, say, within 0.001 degrees, the horizontal error end up with almost 30 feet. So this is not desirable.

Still, we want to get the most out of it. In fact, there are many cases satellite data will help you in the real world. So we are going to see when you should consider using satellite solution.

MARIKO MORIOKA: OK, we have still limitations with the satellite solutions, so let's compare it against other solutions. In this slide, you will see a list of pros and cons when it comes to utilizing satellite, airborne, UAV, and ground survey for each. Naturally, no one single method is superior in all aspects.

In general, you can get an image for the larger area at once if the observer is in the higher observation altitude, but the resolution will be lower. Earth's observation satellites are in around 500 to 700 kilometers, which is 310 to 430 miles, and the highest commercial optical satellite image resolution is 30 centimeters, which is not the greatest solution among others.

Please note that it's important to understand the trade off of each solution, and selecting the best one depends on the project location, target area size, and the project requirements, such as accuracy, cost, and the timeline.

So when a satellite solution works for your project? As I mentioned in the previous slide, you can get larger area at once from a satellite image, but the resolution will be compromised compared to other solutions, and the accuracy will be expected the same. Based on this, a satellite solution works best in a planning phase or operation phase when the resolution and accuracy are not the most critical factors.

Thanks to its shorter lead time, the total cost of managing your project's schedule can be drastically reduced. That's why you should consider adopting a satellite data solution to your project.

So let's take a look at the following two use cases. Both cases, users enjoy the satellite-based geo-data with an Autodesk product. The first use case is from NTC International based in Japan. They are a consulting firm specializing in sustainable development projects all over the world.

In this project, satellite-based data were used for water resource utilization and irrigation drainage development of a dam in Paraguay. Images B, C, and D are captured images showing a cross-section using a 1 meter resolution DTM imported into Civil 3D to calculate the amount of earthworks when the route of the main changes.

The field of survey point, GCP, the ground control point, was also used to improve the positional accuracy of satellite-based DTM and other additional processing done by the data provider so that NTC International could work on analyzing a more precise facility layout and estimating the project cost with the data.

The second case study is from JGC Algeria, one of the biggest oil and energy companies in Algeria and North Africa. They utilize the satellite based one meter resolution DTM, contour line, and also imagery with GCP correction to optimize the pipeline routing.

The contact line was provided in DWG so that the customer could easily work on with an Autodesk software. The satellite-based solution was chosen rather than the classic topographical survey on site because of the tight project schedule and multiple remote locations with no access roads and desert areas. Also, the mobile network does not cover these areas.

Thanks to the wider data coverage, they could handle the changes more efficiently during the project lifecycle they had to deal with. Also, they could save time and costs compared to the traditional topographical survey. That's all for me, and I'm going to pass Sami for the next topic.

SAMI HARADA: So next, let's move on to the topic about Autodesk products, how they handle satellite data and what you should know about it. There are two most important things to handle satellite data. First one is coordinate reference system, CRS. In order to see an object in the right position and the right size, your software needs to understand CRS.

The second one is data interoperability. There are many variations in a single GIS file format. Sometimes you should know the exact format details to avoid an issue in your project.

This diagram shows the level of functionality each Autodesk product offers. In Frameworks, one of the most modern applications, offers the best functionality and capability in terms of handling satellite data. Civil 3D, Maps 3D offer excellent CRS support as well. AutoCAD and AutoCAD LT are limited, but they support a basic and minimum CRS functionality also. There are some good third party solutions to make them work.

And you should also want to consider users who are on viewing environment such as using AutoCAD Plus Object Enabler, DWG TrueView, Autodesk Viewer, and AutoCAD Web or Mobile. [INAUDIBLE] to make your design portable is putting everything in a standard DWG file. Please check this AU class, CES468698, if you are interested in that direction.

So here is a tricky part of Autodesk product portfolio. When you can afford a easy correction, then things can be easy and very straightforward. However, you should aware that managing and maintaining multiple application packages and multiple data files in a different format can be more expensive than you would think.

Civil 3D includes Map 3D functionality, and you can enjoy all excellent GIS-related functionality from them, and everything can be captured within a single DWG, and that is really great.

AutoCAD may sound like a single, simple product, but it depends on where you live, the level of subscription, and/or the version of it. It's different, the fact that it may or may not include Map 3D extension. You should consult your IT manager or reseller if you can access Map 3D functionality from your AutoCAD in case you are not sure about that.

AutoCAD LT is a great application software for people who have a very specific need, but again, it may or may not be available in your region and country.

So I would like to move on the next topic, data compatibility. This table is a list of the result of my experiments with variation of GeoTIFF data and can be read from various Autodesk products.

The point I am trying to make here is not which product is better or which product is worse. What I am trying to say here is there going to be a compatibility issue if you happen to get a wrong data type of satellite data. So please note that results can be different if you are using a different version product or a different platform.

Here is a similar table that describes a compatibility test result with 3D data consumption. Again, do not read this as InfraWorks is the best and AutoCAD is bad. But this might be different with a future release or with a service pack.

So as you might be able to overcome some limitation using a third party solution or workaround such as the information explained in this AU class, so if you are interested in checking.

So next, let's take a look at some tips and tricks when you obtain satellite data. So first of all, pay attention to the resolution. You need at least 2 by 2 pixels to identify an object. The actual effective resolution comes from both sensor resolution and off-nadir angles. So check the ground sample distance, GSD, value when the resolution is critical. For CAD consumption, pansharpened RGB three band imagery in GeoTIFF format would be the best.

And I will explain a little bit more about the very last line of this slide. So this screenshot demonstrates when an image doesn't work as you expected. Because of these off-nadir angles, you cannot see the area behind those tall buildings. Also you may notice that many areas are blackened out due to the shadow or due to the low sun-elevations. If you need a digitized load, for example, in an urban area, pay attention to those factors as well.

OK, here is a list of what you should know when you obtain DEM, digital elevation model. First, decide if you need DSM, DEM, or both. Then CRS, coordinate reference system, should be consistent with other elements of your design.

Vertical datum is also important. Ellipsoidal elevations, that's more popular in a specific field in an industry, but in other fields, it isn't. When the accuracy is very important, consider using GCP to ensure the data you will get will be aligned to the existing data.

OK, then we will take a look at how Autodesk product actually works with some satellite data. I would use [? predict ?] captured static imagery for this recording. Also, I'm not going to show how things work. Instead, I would like to show how things do not work in some cases.

OK, this is what you are going to see when you try to open a wrong GeoTIFF data that are not compatible with your product. As you can see, there are various types of error messages. Some are really obvious and some are somewhat obscure.

So next example, while preparing this presentation, I have seen this kind of unexpected behavior errors. Sometimes AutoCAD and Civil 3D cannot render an imagery correctly, but I have no idea why this happens. Or the thing is, those imagery can totally fine with other software like InfraWorks or other GIS software.

Sometimes I also run into multiple crashes and maybe due to the wrong GS data format, but this not somehow very consistent to reproduce.

So when you consume the DSM from Civil 3D, pay attention to the metadata. Some DEM do not define the NoData value in the metadata in GeoTIFF and Civil 3D does not like it when GeoTIFF is created with BIG-TIFF, which uses 64-bit addresses to support more than 4 gigabyte file size. Civil 3D silently refuses to open it with very little information on what went wrong.

So finally, this is what people [INAUDIBLE] contour lines generated with GIS software or GIS [INAUDIBLE]. Since they are not Autodesk product native, sometimes the clarity of the DWG and DXF data is not perfect. So always run audit command before you start consuming it in your drawing.

MARIKO MORIOKA: Well, we have now come to the end of our presentation for this class. In the next slide, Sami is going to summarize what we talked about.

SAMI HARADA: OK, so this is the summary of what we saw in this class. New type of satellite data are available. So they can be detecting greenhouse gas or geo-hazard. DEM, DSM, DTM, and 3D building data can be generated from satellite data. Although satellite situation is not perfect, people have been using them effectively in their project.

So next point, Autodesk offers a full set of functionalities that are necessary to consume satellite data, CRS support, data compatibility are the two very important key factors. The level of support of each Autodesk product varies, so you must know what you already have, which product you should pick, and how to use it properly.

Lastly, there are many variations in satellite imagery and data, so some can work without any issues, but some cannot. So when it comes to the consuming the geo data, please take a look and be careful.

So all right. Thank you so much for you joining us for this class. We are excited to see how you take those information and use them to respond to the challenges of everything. That brings me to the end of this presentation. Thanks again for your attention. Take care.