Hello, Everyone. I'm not sure where to post this announcement on the Celestia forum. So, I will post here and one other place:
NASA Goddard Space Flight Center is pleased to announce
the immediate availability of open source software for
space mission analysis and trajectory design. The General
Mission Analysis Tool (GMAT) has been under development by
NASA, in partnership with Thinking Systems, Inc., for four
years, and is an ongoing technology development effort.
GMAT is offered free of charge to use, modify, and share as
described under the terms of the NASA Open Source Agreement.
We are excited that this release is the start of a long term
effort to work with the space community to develop and apply
space mission design technology. While GMAT has undergone
extensive testing and is mature software, we consider the
software to be in Beta form. We are in the process of performing
significant testing to prepare GMAT for operational use.
This is a one time announcement. For further information,
please visit the project web site:
http://gmat.gsfc.nasa.gov/.
To receive future announcements, please subscribe to
the project mailing list:
https://lists.sourceforge.net/lists/listinfo/gmat-info
NASA OS Software: The General Mission Analysis Tool (GMAT)
-
Topic authorsteven hughes
- Posts: 12
- Joined: 19.07.2007
- With us: 17 years 4 months
-
Hungry4info
- Posts: 1133
- Joined: 11.09.2005
- With us: 19 years 2 months
- Location: Indiana, United States
-
techno_mage
- Posts: 4
- Joined: 03.06.2007
- With us: 17 years 5 months
very nice ... I started using wxWidgets about a month ago and now suddenly it seems like everyone is using them. I expect IBM will be using them next 
I personally don't get a fast enough framerate in wxWidgets on the Mac or Linux.
Would a GNU-Make makefile be possible within the near future? I could supply one if you want?
Is it possible to visualize the ideal-path zones for space craft?
Maybe it would be worth using dlopen() in celestia's lua implementation to allow for scripting of spaceship pathways.
I am very excited by this discovery, since I have been considering an on-board space craft navigation system, based on graviational fields from simulations ...
I personally don't get a fast enough framerate in wxWidgets on the Mac or Linux.
Would a GNU-Make makefile be possible within the near future? I could supply one if you want?
Is it possible to visualize the ideal-path zones for space craft?
Maybe it would be worth using dlopen() in celestia's lua implementation to allow for scripting of spaceship pathways.
I am very excited by this discovery, since I have been considering an on-board space craft navigation system, based on graviational fields from simulations ...
Thanks for the response. I'm one of GMAT's developers, so Steve asked me to post a reply.
We actually do build GMAT on all three supported platforms (Mac, Windows, Linux) using GNU-Make. The makefiles you see in the source distribution have the suffix ".eclipse" as an artifact -- we used to have separate Mac, Linux and Windows makefiles (with corresponding suffixes), and when I combined all three into a single set of makefiles to ensure consistency, I used the .eclipse suffix to distinguish it from the platform specific versions. (We use Eclipse as our IDE, and build GMAT with the GCC. I do occasionally build from the command line as well using GNU-Make directly.)
You may have noticed that there is a folder in the source distribution named "topLevelBuildFiles" which contains platform specific subfolders. The main differences between the platform build files is contained in the file "BuildEnv.mk", which contains the current build settings, paths to external files, and so forth. I'd certainly prefer that we were using the AutoConf tools so that we could build with "configure - make - make install," but we're just not there yet; hence the somewhat awkward build file structures.
I hope that helps explain the current state of the GMAT build files. I'd like to post this message thread to the gmat-developers list as well. Please let me know if that is a problem.
- Darrel Conway
We actually do build GMAT on all three supported platforms (Mac, Windows, Linux) using GNU-Make. The makefiles you see in the source distribution have the suffix ".eclipse" as an artifact -- we used to have separate Mac, Linux and Windows makefiles (with corresponding suffixes), and when I combined all three into a single set of makefiles to ensure consistency, I used the .eclipse suffix to distinguish it from the platform specific versions. (We use Eclipse as our IDE, and build GMAT with the GCC. I do occasionally build from the command line as well using GNU-Make directly.)
You may have noticed that there is a folder in the source distribution named "topLevelBuildFiles" which contains platform specific subfolders. The main differences between the platform build files is contained in the file "BuildEnv.mk", which contains the current build settings, paths to external files, and so forth. I'd certainly prefer that we were using the AutoConf tools so that we could build with "configure - make - make install," but we're just not there yet; hence the somewhat awkward build file structures.
I hope that helps explain the current state of the GMAT build files. I'd like to post this message thread to the gmat-developers list as well. Please let me know if that is a problem.
- Darrel Conway
I just downloaded GMAT and I'm getting an error about a missing DLL:
libeng.dll
I'm assuming this is part of MATLAB and that shoud mean the following DLLs would be required also:
libmx.dll
libut.dll
Why are they not distributed with the package?
Or do you need to install MATLAB?
I ran the application without MATLAB and it seems to be working perfectly.
I'm wondering if it could be integrated with Celestia. It would be nice to model launches and the deployment of orbital satellites from start to finish in a single environment. If we could find a way to import the scripts then we could open this to a wider audience.
It would also add more realism to Celestia and would add to its role in education too.
libeng.dll
I'm assuming this is part of MATLAB and that shoud mean the following DLLs would be required also:
libmx.dll
libut.dll
Why are they not distributed with the package?
Or do you need to install MATLAB?
I ran the application without MATLAB and it seems to be working perfectly.
I'm wondering if it could be integrated with Celestia. It would be nice to model launches and the deployment of orbital satellites from start to finish in a single environment. If we could find a way to import the scripts then we could open this to a wider audience.
It would also add more realism to Celestia and would add to its role in education too.
A whack in the face with a fish can solve a whole manner of problems.
Fridger,
I agree that it would be best to have only one thread for this topic. Unfortunately, this version of phpbb does not have the ability to merge discussion threads. (Supposedly it will be possible in a future major release.)
The best I can do is lock one of the threads and include a pointer to the other.
So: please use this thread for discussing GMAT. I'll lock the other one, which seems to be wandering off into inappropriate political commentary anyhow.
I agree that it would be best to have only one thread for this topic. Unfortunately, this version of phpbb does not have the ability to merge discussion threads. (Supposedly it will be possible in a future major release.)
The best I can do is lock one of the threads and include a pointer to the other.
So: please use this thread for discussing GMAT. I'll lock the other one, which seems to be wandering off into inappropriate political commentary anyhow.
Selden
-
Topic authorsteven hughes
- Posts: 12
- Joined: 19.07.2007
- With us: 17 years 4 months
Hi DT,
This response is regarding your message above, where you mention needing libmx.dll and libut.dll for one of the executables to run. The GMATBeta executable requires that MATLAB is installed on your machine to run. The libmx.dll and libut.dl libraries are not open source, they are owned by Mathworks. So, we cannot distribute them.
The fact that we have two executables is not something that we like and is something we intend to fix. Once we fix an unnecessary dependency in the code, then we will only have one application file like just about every other software system
If you don't have MATLAB, you can run GMATBetaNoMATLAB. It has the exact same functionality as the GMATBeta exe, except the MATLAB interface, which you can't use if you don't have MATLAB anyway!
Steve
This response is regarding your message above, where you mention needing libmx.dll and libut.dll for one of the executables to run. The GMATBeta executable requires that MATLAB is installed on your machine to run. The libmx.dll and libut.dl libraries are not open source, they are owned by Mathworks. So, we cannot distribute them.
The fact that we have two executables is not something that we like and is something we intend to fix. Once we fix an unnecessary dependency in the code, then we will only have one application file like just about every other software system
If you don't have MATLAB, you can run GMATBetaNoMATLAB. It has the exact same functionality as the GMATBeta exe, except the MATLAB interface, which you can't use if you don't have MATLAB anyway!
Steve
-
ElChristou
- Developer
- Posts: 3776
- Joined: 04.02.2005
- With us: 19 years 9 months
-
Topic authorsteven hughes
- Posts: 12
- Joined: 19.07.2007
- With us: 17 years 4 months
Hi ElChristou,
I posted a GMAT announcement to the Celestia web site because I thought there may be interest. I think it would be great if there is some collaboration between the two projects. But, we'll have to see if there's enough interest. Currently, the GMAT and Celestia systems do not interface with each other.
Steve
I posted a GMAT announcement to the Celestia web site because I thought there may be interest. I think it would be great if there is some collaboration between the two projects. But, we'll have to see if there's enough interest. Currently, the GMAT and Celestia systems do not interface with each other.
Steve
For those of us who are trying to calculate spacecraft trajectories for use with Celestia, this looks like a very interesting tool.
Steve,
For those who aren't on the best of terms with celestial mechanics , programs like this can be intimidating. What's the chance of the documentation including a "worked example"? e.g. takeoff from Earth through landing on Mars?
Steve,
For those who aren't on the best of terms with celestial mechanics
, programs like this can be intimidating. What's the chance of the documentation including a "worked example"? e.g. takeoff from Earth through landing on Mars?Selden
-
ElChristou
- Developer
- Posts: 3776
- Joined: 04.02.2005
- With us: 19 years 9 months
Tx Steve.
But, following Selden and Dirkpitt, wasn't SPICE supposed to do this kind of stuff? (trajectory design and else) (again, sorry if it's a noob question)
Steve, SPICE has been integrated to Celestia because there was an interest by guys from your agency; the job has been done, but as far as what we can see on the forum, to the day, there is a total blackout, no feedback not even talking about an eventual backlash.
Now question; if you talk about an eventual collaboration, what is exactly your idea? what can Nasa guys do for Celestia? (I'm asking because really I'm wondering...)
Or, honestly, will this be another one way collaboration? (sorry to be so direct)
Really I'm wondering what Nasa can offer to such a tiny project...
But, following Selden and Dirkpitt, wasn't SPICE supposed to do this kind of stuff? (trajectory design and else) (again, sorry if it's a noob question)
Steve, SPICE has been integrated to Celestia because there was an interest by guys from your agency; the job has been done, but as far as what we can see on the forum, to the day, there is a total blackout, no feedback not even talking about an eventual backlash.
Now question; if you talk about an eventual collaboration, what is exactly your idea? what can Nasa guys do for Celestia? (I'm asking because really I'm wondering...)
Or, honestly, will this be another one way collaboration? (sorry to be so direct)
Really I'm wondering what Nasa can offer to such a tiny project...
Chris,
SPICE is a data format, similar to but much more detailed than Celesita's rather limited xyz trajectory format. It's used to describe object trajectories and orientations. It's not a program for calculating trajectories. It's an obvious output format for GMAT to generate, although I don't know if it does.
What do you mean by your use of the word "backlash"? That's usually a term used to describe a vehement rejection, which is not what I think you mean.
Just because people don't often discuss Celestia's support for SPICE doesn't mean it isn't being used. There have been several threads here favorably comparing SPICE spacecraft trajectories in Celestia with what NASA has published about the actual trajectories. I know of groups in ESA and NASA who take advantage of that.
After all, how often do you talk about the fetures of the 3D modelling software that you use? Instead, you use them to produce the things that you're really interested in: beautiful 3D models. The tools you use almost become invisible to you. Celestia is such a tool.
One of the collaborations that groups at both NASA and ESA are very interested in is enabling realtime trajectory displays in Celestia, with control by an external program. One of the groups at ESA has almost finalized that code. Hopefully they'll be providing it to Chris (Laurel) for inclusion in the official version of Celestia.
SPICE is a data format, similar to but much more detailed than Celesita's rather limited xyz trajectory format. It's used to describe object trajectories and orientations. It's not a program for calculating trajectories. It's an obvious output format for GMAT to generate, although I don't know if it does.
What do you mean by your use of the word "backlash"? That's usually a term used to describe a vehement rejection, which is not what I think you mean.
Just because people don't often discuss Celestia's support for SPICE doesn't mean it isn't being used. There have been several threads here favorably comparing SPICE spacecraft trajectories in Celestia with what NASA has published about the actual trajectories. I know of groups in ESA and NASA who take advantage of that.
After all, how often do you talk about the fetures of the 3D modelling software that you use? Instead, you use them to produce the things that you're really interested in: beautiful 3D models. The tools you use almost become invisible to you. Celestia is such a tool.
One of the collaborations that groups at both NASA and ESA are very interested in is enabling realtime trajectory displays in Celestia, with control by an external program. One of the groups at ESA has almost finalized that code. Hopefully they'll be providing it to Chris (Laurel) for inclusion in the official version of Celestia.
Selden
-
ElChristou
- Developer
- Posts: 3776
- Joined: 04.02.2005
- With us: 19 years 9 months
selden wrote:Chris,
SPICE is a data format, similar to but much more detailed than Celesita's rather limited xyz trajectory format. It's used to describe object trajectories and orientations. It's not a program for calculating trajectories. It's an obvious output format for GMAT to generate, although I don't know if it does.
Tx for the clarification.
selden wrote:What do you mean by your use of the word "backlash"? That's usually a term used to describe a vehement rejection, which is not what I think you mean.
Yep, sorry, must be "in exchange" or "in return" (used in term of collaboration)
Yep, that's exactely the point.selden wrote:Just because people don't often discuss Celestia's support for SPICE doesn't mean it isn't being used. There have been several threads here favorably comparing SPICE spacecraft trajectories in Celestia with what NASA has published about the actual trajectories. I know of groups in ESA and NASA who take advantage of that.
After all, how often do you talk about the fetures of the 3D modelling software that you use? Instead, you use them to produce the things that you're really interested in: beautiful 3D models. The tools you use almost become invisible to you. Celestia is such a tool.
The tool is here (SPICE), but results from it's use, not really...
So yes it's very nice to implement technologies into Celestia for the use of a few professionals, now my question is what can we expect from those "collaborations"? ... fame? ... else?
selden wrote:One of the collaborations that groups at both NASA and ESA are very interested in is enabling realtime trajectory displays in Celestia, with control by an external program. One of the groups at ESA has almost finalized that code. Hopefully they'll be providing it to Chris (Laurel) for inclusion in the official version of Celestia.
Yep, hopefully... but even in that case, and? what next?
What I mean is that it would be nice to see something out of those collaborations, what, I don't really know, it's why I was asking...
Right now GMAT does not write SPICE formatted data. It can generate a text ephemeris, though -- epoch, X, Y, Z, for example. I suspect that one way we may be able to work together is to have Celestia read and display a GMAT trajectory by importing that type of data. I haven't looked for it yet -- is there a data format spec someplace that I can read to see if it is something we can do pretty quickly?
I saw a few questions about how GMAT and Celestia can work together. Right now GMAT has a fairly basic OpenGL plot, but not anything nearly as beautiful as Celestia. One of the goals of the GMAT project is precision orbital (and attitude) control of one or more spacecraft simultaneously, and maneuver planning to accomplish that control. We don't really need glorious graphics to meet that goal, but it sure is nice to have if possible!
There are quite a few sample "script" files in the GMAT delivery. They provide a pretty good introduction at the "demo" level, and you can tweak them to start to get an understanding of the flight dynamics side of things. One thing that we are looking for is some insight into what type of things will help people start using GMAT. Pleas post any suggestions you folks have to the gmat-users mailing list (gmat-users at lists dot sourceforge dot net); that way the whole GMAT team (all 5 of us) will see them.
I saw a few questions about how GMAT and Celestia can work together. Right now GMAT has a fairly basic OpenGL plot, but not anything nearly as beautiful as Celestia. One of the goals of the GMAT project is precision orbital (and attitude) control of one or more spacecraft simultaneously, and maneuver planning to accomplish that control. We don't really need glorious graphics to meet that goal, but it sure is nice to have if possible!
There are quite a few sample "script" files in the GMAT delivery. They provide a pretty good introduction at the "demo" level, and you can tweak them to start to get an understanding of the flight dynamics side of things. One thing that we are looking for is some insight into what type of things will help people start using GMAT. Pleas post any suggestions you folks have to the gmat-users mailing list (gmat-users at lists dot sourceforge dot net); that way the whole GMAT team (all 5 of us) will see them.
Celestia's native trajectory format (SampledOrbit) is indeed epoch, x,y,z quads, with epoch specified as a Julian date and xyz in km. Celestia v1.5 can specify the trajectory file to be relative to just about any coordinate system. Although the data in the trajectory file can be any resolution, right now Celestia stores the xyz coordinates in single precision. That'll be improved to double precision in the near future.
Ceiestia also can use a file of julian date and quaternions to determine the orientation of the object.
Documentation of the trajectory format is included in the Celestia WikiBook at
http://en.wikibooks.org/wiki/Celestia
Ceiestia also can use a file of julian date and quaternions to determine the orientation of the object.
Documentation of the trajectory format is included in the Celestia WikiBook at
http://en.wikibooks.org/wiki/Celestia
Selden
-
ElChristou
- Developer
- Posts: 3776
- Joined: 04.02.2005
- With us: 19 years 9 months
Hello, Darrel, is this a "fictional" trajectory (test) or something existing?
To the GMAT team (Steve, Darrel and others):
Concerning this collaboration, if Celestia is used to visualize the output of GMAT, and if what Celestia gain is only the fame to be used by Nasa, then would it possible to at least "officialese" this "fame", by for example writing something on the GMAT pages about the reasons for choosing Celestia for this task + a few screenshots, links, acknowledgment?
I'm asking because actually Celestia dev is not in good shape, so eventually this kind of thing could help in attracting more professionals interested in space visualization...
Opinion? possible?
To the GMAT team (Steve, Darrel and others):
Concerning this collaboration, if Celestia is used to visualize the output of GMAT, and if what Celestia gain is only the fame to be used by Nasa, then would it possible to at least "officialese" this "fame", by for example writing something on the GMAT pages about the reasons for choosing Celestia for this task + a few screenshots, links, acknowledgment?
I'm asking because actually Celestia dev is not in good shape, so eventually this kind of thing could help in attracting more professionals interested in space visualization...
Opinion? possible?
It's a fictional orbit. Basically, I took GMAT's "default" mission -- the first one we use to sanity test a new build -- added a couple of maneuvers and some propagation, and wrote the results to a file. That part took about a minute. Then I edited the file to set the epoch correctly -- we use a modified Julian data which is offset from full Julian by 2430000 days, so that we can gain some numerical precision in our epoch data. Finally, I added this fictional spacecraft to Celestia's data files.
I think it makes sense to at least help promote Celestia if we are collaborating. I should mention that I'm not a NASA employee, so I can't really speak to what can be done beyond that; anything I say here is my opinion, and shouldn't be taken as something from NASA.
On a lighter note, my wife and I were watching NCIS last night while I was putting together this screen shot, and, while I was importing things into Celestia, the actors on NCIS were finding Celestia on the victim's computer. An amusing coincidence!
I think it makes sense to at least help promote Celestia if we are collaborating. I should mention that I'm not a NASA employee, so I can't really speak to what can be done beyond that; anything I say here is my opinion, and shouldn't be taken as something from NASA.
On a lighter note, my wife and I were watching NCIS last night while I was putting together this screen shot, and, while I was importing things into Celestia, the actors on NCIS were finding Celestia on the victim's computer. An amusing coincidence!