Urthecast Brings You Earth Images and Videos from the ISS (Video)

Video Urthecast Brings You Earth Images and Videos from the ISS (Video) 16

Posted by Roblimo on Friday July 31, 2015 @05:31PM from the sometimes-you-look-at-the-sky-and-sometimes-the-sky-looks-at-you dept.

Video no longer available.

Most of us probably won't ever visit the International Space Station (ISS) and look down at the Earth (motto: "The only planet we know has beer, so let's not ruin it"). Looking at pictures and videos made by cameras mounted on the ISS is about as close as we're going to get. There's already an ISS HD Earth Viewing Experiment on Ustream, but Urthecast is putting out higher-definition images than what you see on Ustream, and has plans to put out even clearer images and video before long. While Urthecast is likely to accumulate plenty of "oohs" and "aahhs" as it rolls along, according to CEO Scott Larson their real objective is to sell imagery -- and not necessarily just from the visible light band of the overall spectrum -- to industrial and government users. People like us are still invited to look at (and marvel at) lovely images of our planetary home.

NOTE: Today's video is about 4:30 long. If you want to watch and listen to more of Mr. Larson, we have a second "bonus" (Flash) video for you. Or you can read the transcript, which covers both videos.

Timothy Lord for Slashdot: Scott, your company has been putting cameras on the ISS and you have just released a major video stream of three different cities on Earth. Can you talk about the technical challenges that are involved in putting cameras on the ISS?

Scott Larson: So we have the rights to put four cameras on the International Space Station. We have two up there right now on the Russian segment and will be installing two more on the U.S. on the NASA segment going up in late 2017. So yeah space is hard, everything about space is hard and expensive, there's no doubt about it. What we have is two cameras: one is a fixed camera, it's pointed directly down and it takes what's called a medium resolution image. It's kind of an always on, takes an endless panoramic image, if you will, of Earth as the Space Station orbits Earth about 16 times a day. Anything that’s 5 meters big you can see, so buildings roads, rooftops, fields, farms, rivers. It is used for wide area coverage. Agriculture and forestry and so forth would be the main use case.

The other camera is mounted on the pointable arm that we can swivel and hover and hold to direct and target. So if you imagine the Space Station is going over San Jose and one camera is pointed directly down, it sees whatever it sees, roads, building things like that. The other camera we decide we want to put over downtown San Francisco, we can point it forward, point it down, point it back for about 60, 75 seconds as it flies past and then it will go on to the next part, which might be LA, San Diego, Mexico City, whatever else and it will take 150 of those videos a day at 1 meter resolution. Anything that’s 1 meter big you can see--cars, buses, boats, plane, groups of people. And that arm is what's vibration. So Space Station it's big, it's about the size of a football field and you have people living up there in the Space Station. It has what's called pitch, roll and yaw--it has all three. So it's kind of going like this and on top of that you have people literally up there on a treadmill and bouncing around and banging into stuff and hammering stuff and so it's kind of vibrating like this pitch low frequency as well as high frequency vibration. And so if you have a camera and you're about 250 miles away traveling about 18,000 miles an hour (about 5 miles a second) and you're trying to lock it in over one area as it vibrates--it is not easy. It is just absolutely not easy and so to come up with the types of videos that you've seen is quite an achievement.

Slashdot: So in order to counteract that vibration--is it primarily in software? Is it mostly hardware? How do you deal with that?

Scott Larson: Yeah. So it's a combination of things, one is you want the hardware because this arm--we call it a BPP, a biaxial pointing platform. This arm has gears and you want these gears to be entirely smooth so that as this thing goes over and hovers over San Francisco, the Space Station is flying, you want those gears to be smooth. And that's not easy to do to start off with. And so you do as much as you can from a hardware standpoint to make sure that the gears move smooth, that it doesn't jerk, that it doesn't vibrate inside the engine, inside the motor. And then on top of that you do an awful lot of processing to make sure that buildings are smooth, roads lined up and you look at things from different angles. It does come down to processing.

Slashdot: Now, the three cities that we've seen, this resolution has very nice, high-resolution footage from space right now. Is that footage fairly typical? Or is that sort of the best 30 seconds of the day? Is it pretty smooth in its general nature?

Scott Larson: So those videos looked great--Boston, Barcelona, London. Those videos looked great. The way satellites typically work is that you put them up there and you know our camera is on the Space Station so it's a little bit different but generally you launch satellites into space, you spend a bunch of time to fine tune it, there's always technical issues. You have to work your way through that and then you come up with first light, which is the first time you're comfortable unveiling your imagery to the world. We did that a few weeks ago and then you spend the next eight years making it better. And so most satellites are their best. The imagery is the best, the data comes down the best. Everything works the best six months before they splash into the ocean. And so whether you are at 85% of where you want to be whatever the number is you're good enough to start selling hot dogs from the store and then you spend the next number of years getting it better. So those videos we believe will continue to look better and better over time as we improve algorithms.

Slashdot: Can you talk a little bit about the way the data gets from the ISS to the ground--that’s a lot of data to send--how reliable is the connection? How fast is it? How do you characterize it?

Scott Larson: So our two cameras right now are on the Russian segment of the Space Station. Our next two cameras are going to be on the U.S. segment of the Space Station and there are two different downlinks: Russia’s downlink is what's called X-Band downlink, the camera is basically record, record, record they can save it in space; cache, if you will, and then as it goes over an antenna on the ground, the data drops down. So it comes down in chunks effectively. We also have access to streaming, which is the way that NASA gets the data down. They have a Space Station whipping around Earth like this and they have a network of geostationary satellites way out and the data goes from the Space Station out around, around, around and it comes down at Houston, in fact, it's always on. So some of our data comes down using the Russian downlink, it kind of comes down in chunks and we get it from these antennas all around the world. It comes to us we process it and then do whatever we do with it. Other parts of the data come down through other means so we have redundancy from a downlink standpoint. We send down about 200 gigabytes of data a day; so we end up with petabytes in the not too distant future, it’s huge amounts of data.

Slashdot: How do you deal with all the data on the ground? What do you do in terms of, is most of your processing actually done there? I think some is done in camera to save bandwidth.

Scott Larson: So we do some in space. The vast majority we send down and then we decide what do we want to do with it. Is it data? Is it imagery that's useful, relevant? How imagery typically works is that it becomes more and more valuable as we build up an archive. So our plan is to archive all of it effectively as much as we can. There's no sense archiving clouds, I mean, so you throw away some stuff that's useless, but you archive as much of it as you can forever and we store it in AWS.

Slashdot: Now speaking of that, you've put a lot of the data that you've stored public facing – far more than a typical satellite imagery program does. Can you talk about the open source nature -- you’re using some open source software as well – what are some ways that's true?

Scott Larson: Yeah. So our plan, our vision is basically to democratize Earth observation. Which to us means taking a view of space and a dataset that's really only been available to a few people—governments, large corporations, hedge funds--those kind of groups, and opening it up as far and as broadly as possible through an API, and allowing developers to make games, apps, educational tools all based around our geospatial based dataset. Which is our data, data from other cameras, other sensors anywhere we can, any kind of Earth observation imagery that we can cram into an API and open it up, and then allow other people to make whatever they’re going to make out of it.

Slashdot: Now you've also just announced less than two weeks ago plans for a constellation. Can you talk about what does that bring that cameras aboard the ISS can’t do?

Scott Larson: So, our benefit at the core of our business is that we've been able to structure a relationship with an international space station where we give them the cameras, they do the launch, the installation, the power, the downlink and then basically we split the data. On the Russian side they take the data of Russia, and they give it to the ministries of forestry, farming, mapping and so forth, and we take the data of the rest of the world. And so our cost is only the cost of the cameras which is about $30 million or $40 million. It doesn't include the cost of the launch and the installation and the space walks -- we need two spacewalks in the downlink – that was kind of provided under this barter. So because our cost of getting the data down is so exponentially less than what other satellites cost, we can get creative, we can open it up, we can give it away, we can do cool stuff. At the core of Urthecast is to take the vision of Earth observation, get it out as far and broad as possible on the one hand, and secondly, we think that when you look at Earth from space you get changed, and you come back fundamentally different. When you talk to astronauts, and I've talked to eight or nine of them, 10 of them, they all say that going to space changed them. They get up there they see how small Earth is, how fragile, there is no borders or countries, and they come back fundamentally changed. And so our belief is that if we can take that view that astronauts have, open it up, it’ll have an impact. It'll enable an impact because of the view that that you get when you look at Earth and see how small it is, how fragile. So we want to somehow contribute to that a little bit as well.

Slashdot: And with 16 you'll actually be able to cover a larger area of the Earth itself?

Scott Larson: And then the point of the space station is that we only have one, so if something's happening over China and the space station is over San Francisco it could take a couple of hours or perhaps even a couple of days to get back over there. And so with a constellation we have radar and optical. So you can operate those two satellites in tandem. They fly in pairs, the optical one is about a minute behind the radar one, radar you can see through clouds, at night, kind of underground and so forth. We’ll have a cloud camera on the radar satellites, so if it's coming over, let's say, San Francisco and half of the Bay Area is under cloud, and this cloud camera can point out where the pockets of cloud-free area are. It sends an RF signal to the optical satellite behind and if there's 10 targets, look at San Jose, Mountain View, downtown, Oakland whatever it is in that area; we can point the optical camera towards an area that doesn't have cloud cover. And so you can train those two things to work together and you can also put those two datasets together – radar and optical – on top of each other, and end up building up kind of a cloud-free layer of Earth based on radar and optical data which is unique. And then your revisit goes through the roof. You can take a picture of any part in the world three or four times a day as opposed to once every couple of days.

Slashdot: Now, Scott as space tourism has gotten cheaper, do you think that you will be able to actually get a first-person view from space as well as from cameras?

Scott Larson: Myself personally?

Slashdot: That's right.

Scott Larson: There is a fitness test and there's probably an intelligence test. Space tourism is hard – rockets blow up as we know. I think Sarah Brightman was actually supposed to go this year, and then passed on it or it didn’t work out. And so it's – space is pretty cool, there's only a few people in the world that have ever had that privilege, so I'm not sure if I'm going to be the next person up, but it’d be exciting.

This discussion has been archived. No new comments can be posted.