I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the most error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would result in less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out for the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster
I would suggest you go with #1 or #2. The added complexity of method #3 probably won't pay any significant dividends in practical terms. You could always implement #3 for version II. :-)
Will you be considering squint? Frankly, I'm not sure any current satellites are using antennas where squint would play a part.
Regards...
On 03/25/2013 11:15 AM, Joseph Armbruster wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the most error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would result in less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out for the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Further to what Gus writes, I think Method #3 will suffer from two assumptions, giving an impression of precision when less should be expected. First, you are highlighting the shadows that mountains and other terrain should give, but which are only applicable to visible light. Radio waves bend and knife-edge diffract over and around these things, so you're eliminating areas from being in serviceable view when they could be interesting to try, if not perfectly useful.
Second, local obstructions such as buildings, trees, and other stuff that aren't represented on Google Earth can be a big factor in the success of any satellite pass, especially if / when we ever get back some microwave capability in orbit. I have a small video camera mounted on my Az/El rotor boom because I have this huge oak tree immediately behind my house, and it was critical to know where it was - one large limb in particular - compared to AO-40, in order to make a contact in that direction. That tree is big, but I suspect not sufficient to show up on Google Earth.
My recommendation is that you go with simple and easy now, and update it later when you need something with more precision, for example, if / when we get a high orbit target to aim precisely at. Maybe we'll have a Google Backyard View by then.
Greg KO6TH
Gus wrote:
I would suggest you go with #1 or #2. The added complexity of method #3 probably won't pay any significant dividends in practical terms. You could always implement #3 for version II. :-)
Will you be considering squint? Frankly, I'm not sure any current satellites are using antennas where squint would play a part.
Regards...
On 03/25/2013 11:15 AM, Joseph Armbruster wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the
most error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would
result in less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out
for the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Greg,
You bring up good points across-the-board. Note: I made it a point to explicitly mention the lack-of-rf-considerations in my original email.
You mentioned Google Backyard View.... what's interesting about my implementation is that if you zoom-in to the ground in Google Earth, you enter "Ground Level View". If you look up into the sky, you can actually see the paths that i'm rendering. I could potentially style the orbits based on the different views. From the ground-view perspective, I could display the AZ / EL. So if someone new to the hobby wanted to make sure they were pointing the right way, all they would have to do is look.
There are all sorts of possibilities when it comes to having a Google-Earth like interface. It is extremely trivial to generate description bubbles and style them cleanly. For our next satellite, we could auto-generate placemarks of the telemetry on the orbit in near-real-time, so when people fly Google Earth, they could see the spots immediately. I think the whole thing is just fun :-)
Joseph Armbruster
On Mar 26, 2013, at 12:49 AM, Greg D wrote:
Further to what Gus writes, I think Method #3 will suffer from two assumptions, giving an impression of precision when less should be expected. First, you are highlighting the shadows that mountains and other terrain should give, but which are only applicable to visible light. Radio waves bend and knife-edge diffract over and around these things, so you're eliminating areas from being in serviceable view when they could be interesting to try, if not perfectly useful.
Second, local obstructions such as buildings, trees, and other stuff that aren't represented on Google Earth can be a big factor in the success of any satellite pass, especially if / when we ever get back some microwave capability in orbit. I have a small video camera mounted on my Az/El rotor boom because I have this huge oak tree immediately behind my house, and it was critical to know where it was - one large limb in particular - compared to AO-40, in order to make a contact in that direction. That tree is big, but I suspect not sufficient to show up on Google Earth.
My recommendation is that you go with simple and easy now, and update it later when you need something with more precision, for example, if / when we get a high orbit target to aim precisely at. Maybe we'll have a Google Backyard View by then.
Greg KO6TH
Gus wrote:
I would suggest you go with #1 or #2. The added complexity of method #3 probably won't pay any significant dividends in practical terms. You could always implement #3 for version II. :-)
Will you be considering squint? Frankly, I'm not sure any current satellites are using antennas where squint would play a part.
Regards...
On 03/25/2013 11:15 AM, Joseph Armbruster wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the most error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would result in less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out for the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
There is so much variability in the footprint that #1 or #2 would probably suffice. Working satellites on the fringes of AOS and LOS wheigh heavily on your location, surroundings, and equipments that it's impractical, it's just a visual "references" that shows relative satellite reception.
Hope this helps.
- Brent, KB1QD
On Mon, Mar 25, 2013 at 11:15 AM, Joseph Armbruster < josepharmbruster@gmail.com> wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the most
error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would result in
less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out for
the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
On Mar 25, 2013, at 8:15 AM, Joseph Armbruster josepharmbruster@gmail.com wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker.
InstantTrack uses a spherical Earth model for ground footprints.
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
I believe that local terrain is dominant over any large-scale terrain effect you'd be able to see on a footprint map. It would be worth modeling for AOS/LOS or mutual visibility purposes.
73 -Paul kb5mu@amsat.org
My humble suggestion:
1. Implement option 1 for the satellite footprint. 2. If you decide to give the users the ability to input their location, them the option to provide either a single minimum elevation angle or a local map -- i.e., 360 individual minimum elevations as a function of Azimuth. It's much easier to project this and the user is generally interested in an unobstructed LOS with respect to his/her location.
73, Ken N2WWD
Sent from my iPad
On Mar 25, 2013, at 11:15 AM, Joseph Armbruster josepharmbruster@gmail.com wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the most error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would result in less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out for the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
On 3/25/2013 6:42 PM, Ken Ernandes wrote:
- If you decide to give the users the ability to input their location, them the option to provide either a single minimum elevation angle or a local map -- i.e., 360 individual minimum elevations as a function of Azimuth. It's much easier to project this and the user is generally interested in an unobstructed LOS with respect to his/her location.
It's not the best resolution but in the image below, you can see how there are "cut outs" in the circles surrounding NASA's ground stations - the software has clearly implemented the idea Ken outlined above. For example, there is apparently some obstruction to the south east of the Hawaiian tracking station. If the sub-satelite point is inside the white line it's AOS. The surface of the earth visible to the shuttle, on the other hand, is simply a red circle, just faintly visible in this image.
http://vault.newsfromspace.com/missions/sts114/STS114_land-5.jpg
-Joe KM1P
Joe,
What tracking program is that?
I was thinking of making my satellite icons configurable. This way the satellite would look like the actual satellite. Even better, I could store collada models for them and load them in place of the icon. Oh la la, that's an idea.
Joseph Armbruster
On Mar 26, 2013, at 7:58 AM, Joe Fitzgerald wrote:
On 3/25/2013 6:42 PM, Ken Ernandes wrote:
- If you decide to give the users the ability to input their location, them the option to provide either a single minimum elevation angle or a local map -- i.e., 360 individual minimum elevations as a function of Azimuth. It's much easier to project this and the user is generally interested in an unobstructed LOS with respect to his/her location.
It's not the best resolution but in the image below, you can see how there are "cut outs" in the circles surrounding NASA's ground stations - the software has clearly implemented the idea Ken outlined above. For example, there is apparently some obstruction to the south east of the Hawaiian tracking station. If the sub-satelite point is inside the white line it's AOS. The surface of the earth visible to the shuttle, on the other hand, is simply a red circle, just faintly visible in this image.
http://vault.newsfromspace.com/missions/sts114/STS114_land-5.jpg
-Joe KM1P _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Joe -
I did the implementation described in a professional flight dynamics software program. The convention (as specified by the gov't) is to have an array of 360 azimuthal "obscura" elevations. The first entry defines the minimum elevation for 0 to 1 deg in azimuth, the second entry defines the minimum elevation for 1 to 2 deg in azimuth, up to the 360th entry defining 359 to 0/360 deg in azimuth. The 1-deg wide azimuthal increments are narrow enough that you get an excellent obscura picture even with the elevations being a step function. It provides a great way to anticipate when obstructions may interfere with an RF or visible LOS.
73, Ken N2WWD
Sent from my iPad
On Mar 26, 2013, at 2:34 PM, Joseph Armbruster josepharmbruster@gmail.com wrote:
Joe,
What tracking program is that?
I was thinking of making my satellite icons configurable. This way the satellite would look like the actual satellite. Even better, I could store collada models for them and load them in place of the icon. Oh la la, that's an idea.
Joseph Armbruster
On Mar 26, 2013, at 7:58 AM, Joe Fitzgerald wrote:
On 3/25/2013 6:42 PM, Ken Ernandes wrote:
- If you decide to give the users the ability to input their location, them the option to provide either a single minimum elevation angle or a local map -- i.e., 360 individual minimum elevations as a function of Azimuth. It's much easier to project this and the user is generally interested in an unobstructed LOS with respect to his/her location.
It's not the best resolution but in the image below, you can see how there are "cut outs" in the circles surrounding NASA's ground stations - the software has clearly implemented the idea Ken outlined above. For example, there is apparently some obstruction to the south east of the Hawaiian tracking station. If the sub-satelite point is inside the white line it's AOS. The surface of the earth visible to the shuttle, on the other hand, is simply a red circle, just faintly visible in this image.
http://vault.newsfromspace.com/missions/sts114/STS114_land-5.jpg
-Joe KM1P _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
On 3/26/2013 2:34 PM, Joseph Armbruster wrote:
Joe,
What tracking program is that?
I am not sure what they call it, but I believe it is a custom NASA program. It was used in the Shuttle mission control center, and they often put it up on NASA TV when S band telemetry was available, but Ku band TV was not.
-Joe
--- On Wed, 3/27/13, Joe Fitzgerald jfitzgerald@alum.wpi.edu wrote:
On 3/26/2013 2:34 PM, Joseph Armbruster wrote:
Joe,
What tracking program is that?
I am not sure what they call it, but I believe it is a custom NASA program. It was used in the Shuttle mission control center, and they often put it up on NASA TV when S band telemetry was available, but Ku band TV was not.
-Joe
It *MIGHT* be (or, at least at one time was) a program called DEMOS (Distributed Earth Model and Orbiter System) running on a Sun workstation.
73, de John, KD2BD
-- Visit John on the Web at:
Ken,
I have already implemented the concept of ground station, albeit, i'm not sure I like the way I have the configuration file set up, see:
ground station implementation: Google Earth Satellite Tracker - Ground Stations U... los implementation: Google Earth Satellite Tracker - Line of Sight Upd...
I'm likely going to implement 1 and move on for now. With respects to the ground station, I like the idea of having a minimum elevation angle, that would be insanely easy to implement. Expect these two to be implemented later tonight :-)
Joseph Armbruster
On Mar 25, 2013, at 6:42 PM, Ken Ernandes wrote:
My humble suggestion:
- Implement option 1 for the satellite footprint.
- If you decide to give the users the ability to input their location, them the option to provide either a single minimum elevation angle or a local map -- i.e., 360 individual minimum elevations as a function of Azimuth. It's much easier to project this and the user is generally interested in an unobstructed LOS with respect to his/her location.
73, Ken N2WWD
Sent from my iPad
On Mar 25, 2013, at 11:15 AM, Joseph Armbruster josepharmbruster@gmail.com wrote:
I can not decide how to implement ground footprints with my google earth satellite tracker. I figured, since I can't make up my mind, I should get a second (and third, and fourth) opinion. For this thread, I would like to discuss how satellite ground-footprints should be implemented. A quick brainstorm led me to three possible implementations (I am leaning towards 3). For each of these, I assume that a geographic line-of-sight footprint is desired with no RF characteristics taken into consideration:
option 1 : assume a spherical earth model and project a polygon downwards towards the footprint
- note: this is obviously the easiest approach but will result in the most error
option 2 : assume an ellipsoidal earth model and project an irregularly shaped polygon downwards towards the footprint
- note: this is arguably more difficult than option 1 and would result in less error
option 3 : use a digital elevation model and an ellipsoidal model to cull-out regions that are not visible due to geographic features and project an irregularly shaped polygon downwards towards the footprint
- note: In this case, our footprint polygon would have holes cut out for the regions that are culled out by mountain ranges, canyons / etc... Obviously, this would be the most difficult to implement but would likely be the best visual representation. The problem is, I would never dream of distributing DEMs for the entire Earth with my tool, even DTED0 would be absurd in my opinion. I could make the elevation queries accessible using a web-service, but then the user would be tied to the internet. The other option would be to allow the users to download their elevation data into a cache, then the tool would just load / use it. This way the user would only have to obtain the elevation data for their region of interest. Maybe that would be the best approach? I am open to suggestions!
If you have any experience visualizing footprints, please let me know. I would be interested in hearing your lessons-learned. These are what the line-of-sight indicators look like right now: Google Earth Satellite Tracker - Line of Sight Update
I am open to comments and suggestions, Joseph Armbruster _______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
participants (8)
-
Brenton Salmi -
Greg D -
Gus -
Joe Fitzgerald -
John Magliacane -
Joseph Armbruster -
Ken Ernandes -
Paul Williamson