Michael Turner <michael.eugene.turner@gmail.com> a écrit :
I'm involved in a project for the 2015 International Space Apps Challenge.
I'm listing both Orekit and Rugged as possible resources. Orekit proved
useful (thanks to the help of the Orekit team) in winning a prize at the
Tokyo Space Apps Challenge in 2014.
Hi Michael,
http://tokyo.spaceappschallenge.org/2014/img/release/PressRelease_H_20140423_02.pdf
This was for the project Sprite Orbits
https://github.com/ProjectPersephone/SpriteOrbits
This year, I want to be involved in something more related to remote
sensing. The premise is remote sensing from equatorial LEO satellites, so
that visitation rates are high, permitting /potentially/ meaningful
oversampling if the effects of transitory shadows can be filtered out.
https://github.com/ProjectPersephone/Moombi
I'm very new to remote sensing and satellite image analysis. It seems to me
that, for /some/ of what Moombi is about (flood and landslides, and their
ecological effects), Rugged might be ideal. Landslides can cause slight
changes in terrain, and more significant changes in coloration. Flooding
might be considered similarly. I'd like to get some opinions on this. Am I
on the right track?
Perhaps, Rugged may help, I am not sure. One of the main input of Rugged
is a Digital Elevation Model, and the library will compute geometric correction
grids that will be fed into an image processing chain to compute ortho-rectification
(another open-source library, Orfeo Toolbox <http://www.orfeo-toolbox.org> may be
helpful there. Digital Elevation Models accuracy is often in the range 10m-20m (see
<http://geodesy.curtin.edu.au/local/docs/Rexer_Hirt_2014_DEM_AJES_av.pdf>. We
have already used Rugged with both SRTM (using a link to the GDAL library to
load the DEM) and ASTER (using a custom loader I can provide to you if you want,
it is in the history of the git repository).
There are a few orbital dynamics issues in this -- for example, it would be
nice to model a whole ELEO remote-sensing satellite mission from launch to
reentry, maybe even the spacecraft orientation problem. But I think terrain
mapping has to be where we concentrate. I just want to be sure that Rugged
can be useful to us. If it is, well, the Apps Challenge is the beginning,
not the end, for the Moombi project.
In any case, I think that your needs would also involve mostly image
processing, the terrain modeling would be just one step of the process.
I agree the Orekit/Rugged pair is a good starting point, mainly as Orekit can
simulate accurate meta-data (orbit, attitude) to provide to Rugged and build
the full geometric model. You do not need to have a complex set of applications
or already built reference data: this helps a lot during initial definition and
design. As per version 7.0, Orekit also provides the DSST semi-analytical propagator
that can be used for very long term propagation, up to full spacecraft lifetime.
For ELEO orbits, I guess you will not get too much out-of-plane perturbations,
so a simple periodic semi-major axis with a pair of tangential maneuvers to
compensate for atmospheric drag would probably be sufficient if you intend to
have a spacecraft that has thrusters for station-keeping. This can be done
easily as our implementation of DSST fulfill all the requirements of Orekit
propagators and hence supports both events (to trigger maneuvers when altitude
goes below a specified minimum) and impulsive maneuvers triggered by such events.
best regards,
Luc
Regards,
Michael Turner
Executive Director
Project Persephone
K-1 bldg 3F
7-2-6 Nishishinjuku
Shinjuku-ku Tokyo 160-0023
Tel: +81 (3) 6890-1140
Fax: +81 (3) 6890-1158
Mobile: +81 (90) 5203-8682
turner@projectpersephone.org
http://www.projectpersephone.org/
"Love does not consist in gazing at each other, but in looking outward
together in the same direction." -- Antoine de Saint-Exupéry