4th rock from the sun
Nov 21 2014, 03:13 PM
Malmer, I never meant to downplay your work.
I like your results and they are interesting. You already have Philae coming within meters of those cliffs where it landed and those are great results.
My guess it that some higher topography + different angle would give you a landing at the right spot.
But if you look at the discussion, it was drifting into alternate sites and speculation and not on how to better your results. I commented on that.
Malmer
Nov 21 2014, 03:27 PM
oh... sorry I misunderstood. Thank you for the nice words.
If I where even a 100th as proficient in human to human interaction as in image processing I would have really had something going for me...
chemman
Nov 21 2014, 03:27 PM
QUOTE (4th rock from the sun @ Nov 21 2014, 09:23 AM)
Check Post #761 for the landing site location.
I mentioned it before, all the official / semi official information gives that spot as the general landing location.
Yes, it's on the "crater" rim and that's consistent with the surface panorama.
Don't see much reason for coming up with alternate possible sites.
I wasn't trying to say there needed to be a different site, I was just trying to rationalize the ESA predicted location with the general post bounce direction and Malmers sim. I was thinking either the bounce angle needed to be higher to get more hang time and curve or it may have had another direction change due to impact with the crater rim. Probably the former than the latter.
4th rock from the sun
Nov 21 2014, 04:19 PM
All is well.
Never extrapolate too much from of forum comments.
They are just words with little context and not everybody is a native english speaker.
Non-natives (my case) may do a poor choice of words and not come across as nice or friendly as they intended.
Just a few on topic comments
A few posts back Orbiter space sim came up. I'm a big fan of Orbiter and do some addons for it. It's a realistic software for things like orbital maneuvers, gravitational slingshots, fuel consumption, etc. Those are as realistic as you can get out of a free simulator.
But for surface landings, it not detailed enough. The surface is always at datum and the gravity is point source.
fredk
Nov 21 2014, 04:31 PM
QUOTE (Malmer @ Nov 21 2014, 02:56 PM)
Well this is very interesting. The new plot shows a yellow diamond-shaped region, but doesn't say what it represents, as well as narrower regions. My guess is the larger diamond is the lander region based on measurements through the core, while the narrower regions are from direct line of sight. But that's just a guess.
But the diamond region has shifted south (or down on the images, at least) considerably from the previous estimate (blue diamond on previous map), which puts it much closer to where Malmer's simulated lander flew over! We can even imagine that the diamond will shift again, although the overlap between yellow and the smaller regions is reassuring.
The other thing I notice is that there are two candidate landing areas (blue strip and green ellipse), based on two different nucleus shape models. We have to take the disparity between the green and blue regions as an indication of the effect of the shape model uncertainty on the position determination. In reality, we surely can't say that either one or the other is correct, and something in between or even somewhat beyond might be closer to the truth. So the lander may be anywhere between the blue and the green and even somewhat beyond.
fredk
Nov 21 2014, 04:56 PM
QUOTE (jmknapp @ Nov 21 2014, 11:25 AM)
The axis points to RA 69.4160° dec 63.9879° in the J2000 equatorial frame--assuming reality matches the model. Relative to the ecliptic (SPICE frame ECLIPJ2000) the axis points to RA 78.1392°, dec 41.3938°.
Technically when you specify coordinates relative to the ecliptic you don't call them RA and Dec. They're called longitude and latitude and given symbols l and b (for heliocentric ecliptic coordinates; greek for geocentric). RA and Dec are reserved for equatorial coordinates.
But the good news is that your numbers agree with one of the two references posted above.
MahFL
Nov 21 2014, 05:24 PM
scalbers
Nov 21 2014, 06:15 PM
Nice to see 4th rock's post #793 with a data point that seems to indicate the sun's location near the upper right corner of the 360 degree cylindrical projection (1st image). If we know the time of the panorama, then we'd have the cometary declination, right ascension, and hour angle of that spot. If additional images will be available at other times, then we can add more points.
As per chemman, it will be interesting in the simulations to see how the gravity, rotation, horizontal/vertical trajectory, shape model and flight time can all come together to come up with the actual landing spot.
jgoldader
Nov 21 2014, 06:40 PM
I'd heard about a limited ability of Philae to "hop," but haven't seen any confirmation of this, nor have I seen any technical paper about the landing gear. (Harpoons, yes, but not the gear itself, from what I can recall.) Vigorous searches haven't turned up anything, either.
Can someone here enlighten us? Thanks!
surbiton
Nov 21 2014, 06:51 PM
QUOTE (Malmer @ Nov 21 2014, 02:56 PM)
The new site is definitely below and to the left [ South ? ] of the "old" blue diamond. However, the lie appears to be "better" than the blue diamond.
Unless, Philae is actually in a hole, the lighting conditions ought to be better.
BTW, I am no scientist, just an enthisiast !
jmknapp
Nov 21 2014, 06:51 PM
Some of the orbital maneuvers around landing and after:
Click to view attachmentIt's kind of like Rosetta "threw" the lander at the comet (the segment between 6:14 and 9:15 on the 12th) & from there it was free fall, since the ADS (active descent thrusters) had failed.
dvandorn
Nov 21 2014, 07:15 PM
Yes, Philae was in free fall, but the gravity of C-P is so small that Philae was primarily following a straight line from its point of release, with a very minor perturbation induced by C-P's low gravity.
Which makes me wonder exactly how much velocity was imparted to Philae during the bounce. Philae was slowly drifting towards the comet, while it seems the comet was moving faster than Philae was in terms of rotation. The contact with the comet had to have imparted velocity to Philae, the vector of that velocity defined by the rotational energy of the comet.
It was obviously not enough to exceed orbital velocity for C-P, but it may have been close. Or perhaps it would have been orbital velocity had Philae not relied on some lithobraking for its final stop. I would be interested in seeing just how much velocity was imparted, and what Philae's resulting vector was. We should be able to backtrack that once we have the exact landing site, but it's also possible to predict with good accuracy the landing site from those data.
-the other Doug (With my shield, not yet upon it)
djellison
Nov 21 2014, 07:15 PM
QUOTE (jmknapp @ Nov 21 2014, 10:51 AM)
It's kind of like Rosetta "threw" the lander at the comet (the segment between 6:14 and 9:15 on the 12th) & from there it was free fall, since the ADS (active descent thrusters) had failed.
That was the nominal plan from the beginning. Moreover the ADS was not planned ( even if it had been operational ) to be used during the descent - only on touchdown.
jmknapp
Nov 21 2014, 07:34 PM
It's a document with a lot of placeholders, but the 2003
Rosetta Lander User Manual has a section:
QUOTE
2.3 Lander Subsystems
2.3.1 Active Descent System
An Active Descent System (ADS) provides vertical acceleration during descent and immediately after touch down as hold down thrust.
djellison
Nov 21 2014, 07:46 PM
And indeed - it could have been used during descent, particularly in the case of visiting 46P, not 67P. But for 67P it was NOT planned to be used during descent - they so said several times during pre-landing briefings.
Explorer1
Nov 21 2014, 08:08 PM
Considering that Wirtanen's gravity is even weaker than 67P, it's probably a very good thing that the launch was delayed; the bounce might have reached escape velocity there...
Marvin
Nov 21 2014, 08:34 PM
I wonder if Philae drifted across the crater until it hit the crater wall?
Then it drifted slowly downward until it came to rest in the green area.
Brian Lynch
Nov 21 2014, 11:02 PM
QUOTE (dvandorn @ Nov 21 2014, 02:15 PM)
... I would be interested in seeing just how much velocity was imparted, and what Philae's resulting vector was.
Attached is the speed of Philae during descent with respect to the centre of 67P/C-G (note that this is from the predicted trajectory, which was very closely followed during the actual descent). Also, forgive the time axis, it is referenced to some arbitrary time I chose for the start of the data extraction.
Rosetta starts at an initial orbit before making a sudden 90 deg turn onto a collision course with the comet, where it releases Philae shortly thereafter (the lander actually loses some speed from the velocity change due to the mechanical separation system). Rosetta then diverts to a safe orbit while Philae free falls to the surface. You can see the acceleration due to 67P/C-G's gravity here as well, which is not insignificant. After touchdown, the speed drops sharply from 1.0338 m/s to 0.3215 m/s.
The second attachment shows the velocity vectors just before (
blue) and just after (
red) touchdown (again, with respect to the comet centroid). Since the nominal trajectory assumed the vehicle would be anchored, the red vector is the velocity vector for the comet surface at the landing site.
Jorn Barger
Nov 21 2014, 11:15 PM
QUOTE (Marvin @ Nov 21 2014, 02:34 PM)
I wonder if Philae drifted across the crater until it hit the crater wall?
i suggested this in post 406: "isn't the most likely scenario that its lateral motion was stopped by a cliff, so the direction it was coming from-- where the original landingplace was-- should be comparatively clear, and the opposite side much steeper?"
fredk
Nov 22 2014, 12:05 AM
QUOTE (Brian Lynch @ Nov 21 2014, 11:02 PM)
Attached is the speed of Philae during descent with respect to the centre of 67P/C-G
This is nice. So it looks like before landing, the nominal velocity was close to the x-y plane, and after close to the x-z plane? How are the coordinates defined - is y roughly orthogonal to the surface? Can you relate x and z to directions on the orbital views of the landing area?
What is the vector change in nominal velocity components parallel to the surface just before and after landing? If we could relate this to the orbital views, it would tell us what horizontal component the lander had, relative to the ground, just prior to bounce. That would be very interesting to compare with the observed bounce direction. It's hard to judge this from the pre-landing OSIRIS views due to projection effects.
Brian Lynch
Nov 22 2014, 12:14 AM
The data was exported in the J2000 frame (Earth centered inertial frame) for the spacecraft and the comet, and then I am subtracting the two to get Philae and Rosetta's trajectories with respect to the comet's (but still in an inertial frame).
If we could define the comet attitude and spin with respect to the 3D model coordinate frame (and get the scale correct) then all of those interesting questions could be answered! Unfortunately I have not finished working that out. Perhaps Mattias has some insight based on his analysis, which is far more complete than mine.
jmknapp
Nov 22 2014, 01:25 AM
I suppose the flywheel would complicate the response after the lander contacted the moving surface--the gyroscopic force being 90 degrees to applied force.
Brian Lynch
Nov 22 2014, 01:58 AM
QUOTE (jmknapp @ Nov 21 2014, 08:25 PM)
I suppose the flywheel would complicate the response after the lander contacted the moving surface--the gyroscopic force being 90 degrees to applied force.
Not sure I can confidently say no to that, but keep in mind that any reaction force on the lander would be transmitted to the wheel through its bearings. Even if Philae itself were spinning, which would cause tumbling to occur with that 90 deg offset, the translational motion would not be affected since the reaction on the centre-of-mass is still the same (although the reaction with the ground would itself be affected by any initial spin... as any golfer knows!).
Gerald
Nov 22 2014, 02:10 AM
Torque-induced precession should be applicable during touchdown contact, changing the spin axis for the consecutive ballistic trajectory.
Jam Butty
Nov 22 2014, 01:45 PM
If one looks at the two ROLIS images taken close to the ground (the official one and the leaked one) there is not much evidence of horizontal drift in the landers approach trajectory, so the lander must have picked up its horizontal velocity during the landing itself.
The way I see it the lander came down straight and square as planned, however on landing the leftmost foot/strut (A) hit a rock while the other two feet (B & C) continued on into the soft dust. This would have pitched the lander into a roll and finally a bounce at (D) as one of the legs arced over and dug into the surface again. The flywheel probably prevented the lander from rolling right over onto its 'head'.
Annotated ROLIS image with an OSIRIS post-landing overlay.
Click to view attachmentAnimated ROLIS image with an OSIRIS post-landing overlay.
Click to view attachment
jmknapp
Nov 22 2014, 02:39 PM
A video simulation of the current sunlight situation on 67P over several rotations, using the ESA shape model:
https://www.youtube.com/watch?v=5WgMXvUO-Og...eature=youtu.beRed dot marks the approximate location of the lander. This is assuming it reached the blue stripe area over the ridge as shown in the CONSERT image. Sun declination at 37 degrees N.
fredk
Nov 22 2014, 03:45 PM
Nice. Can you advance several months towards perihelion to see how the lighting would change by then?
Of course this would only be part of the story, since if we're in a hole/crevasse it's hard to know how the lighting onto the solar panels will change.
I'll also add some numbers. After landing we had 1.5 hrs of sun out of a total possible of 6 hr/day. Had we landed on a flat region and had the full 6 hr, that should've been enough to keep the battery charged. By perihelion, the distance to the sun will be less than half the current distance, so the intensity of sunlight will be more than four times higher. So
if we continue to receive 1.5 hrs of sun per day, near perihelion that would be the equivalent of more than 6 hr/day now, in other words, enough to charge the battery!
Of course the crucial question here is how many hours of sun will we get near perihelion. The shadows of the cliffs/boulders/whatever may be worse then, in which case we might not have enough energy to charge.
Finally I'll add this this tantalizing comment from the
Q&A:QUOTE
our first guess is that we'll have enough energy to boot in February and enough to have a link about June. The question we're fighting with (among many others) right now, is how low will the temperatures go in the mean time and what instruments will have survived these low temperatures? - VLL
jmknapp
Nov 22 2014, 05:57 PM
Here's an animation for when the sun reaches the planet's equator (equinox) in about May 2015:
https://www.youtube.com/watch?v=LZ_DOC8aqXEAnd for perihelion:
https://www.youtube.com/watch?v=YfLVFrknSXUAt perihelion the sun is at 50 degrees south declination and the rounder, smoother side of the comet is most exposed to the sun. So does the sun tend to ablate that side more or less evenly? Seems contrary to what they were saying that the notch is possibly caused by more erosion there.
Siman
Nov 22 2014, 06:28 PM
scalbers
Nov 22 2014, 07:05 PM
Nice animations from jmknapp. Looks like the equinox case has the sun transiting near the zenith as I guess Philae is fairly near the cometographic equator.
In post #793 (first image), does the fan shaped flare in the CIVA image pointing to the sky help show the direction towards the sun? It does appear consistent with the shadowing clues. The sun would be near the top of the image projection on the left/right edges. If the camera optics are well understood this would also indicate the angular distance to the sun. Or is the flare a comet jet?
fredk
Nov 22 2014, 08:07 PM
I don't think we've seen much activity this far from the neck. Also, we had very similar looking lens flare in the
post-separation image.I'd guess that the flare would point towards the sun. We also can identify some pieces of the cliff and their shadows in the sunlit frames. And we know the directions of each frame relative to the lander. So if we extend great circles along the flare direction and the shadow directions they should cross close to where the sun was. Knowing the actual (RA, Dec) position of the sun at that time, that would tell us the actual orientation of the lander.
You'd need some software to place the images on a sphere and draw great circles. And the errors in determining the flare and shadow directions may be large. But this could still tell us something very interesting...
scalbers
Nov 22 2014, 08:52 PM
Yes this would be an interesting exercise, and software would help with the great circles, that could conversely be drawn as curves on 4th rock's cylindrical projection. Simply judging by eye, if the lens flare points downward toward the sun relatively nearby, it would be inconsistent with the rightmost image as the sun looks to be above that cliff, as I gather when looking at the bright outlines on the otherwise shadowed ledge. Rather this may be an unusual type of lens flare that converges when going away from the sun.
It seems we'd need two solar fixes at different times to fully solve for the orientation.
Another thing to play with is to estimate a rotation to the local comet horizon and apply it to the cylindrical image.
chemman
Nov 23 2014, 02:54 AM
In the reddit discussion they gave the following info on the general orientation of panel 2 which the one facing out.
"Oh, and we're probably facing NWish with panel 2 since we see the sunset (the sun rises in the East, sets in the West goes over North at noon). - VLL"
Malmer
Nov 23 2014, 12:31 PM
Where is the sun in the post separation image? that would help nail down sun direction...
QUOTE (fredk @ Nov 22 2014, 09:07 PM)
I don't think we've seen much activity this far from the neck. Also, we had very similar looking lens flare in the
post-separation image.I'd guess that the flare would point towards the sun. We also can identify some pieces of the cliff and their shadows in the sunlit frames. And we know the directions of each frame relative to the lander. So if we extend great circles along the flare direction and the shadow directions they should cross close to where the sun was. Knowing the actual (RA, Dec) position of the sun at that time, that would tell us the actual orientation of the lander.
You'd need some software to place the images on a sphere and draw great circles. And the errors in determining the flare and shadow directions may be large. But this could still tell us something very interesting...
scalbers
Nov 23 2014, 02:56 PM
Judging from the lighting on the spacecraft parts the sun seems to be just above the frame in the post separation image. Perhaps that angle can be quantified from spacecraft orbital (SPICE) data as the flare looks rather similar to the post landing image.
Following another clue, which CIVA camera/image would correspond to the number #2 solar panel being able to see the sunset in the NW? I'll assume it's simply the #2 camera.
Question for 4th rock, what would be the basis for vertically shifting some of the images in the post #793 cylindrical mosaic? I think the ideal would be to have a cylindrical projection where the "equator" represents the horizon relative to the sides of the spacecraft.
Emily's post has the images on a more consistent level, though I'm unsure what type of polar map projection it is. Perhaps a guess can be made on that one where the "equator" angle would be? It may have been posted what range of elevation angles CIVA can see.
By the way as I mentioned earlier the perihelion sunlight intensity will be about 6 times that at landing.
Siman
Nov 23 2014, 04:27 PM
Philae’s cameras are looking down, so horizon would be near the top of every frame. You can see it here:
http://www.youtube.com/watch?v=k1IFU6kxcD8
Malmer
Nov 23 2014, 04:45 PM
QUOTE (scalbers @ Nov 23 2014, 03:56 PM)
Judging from the lighting on the spacecraft parts the sun seems to be just above the frame in the post separation image. Perhaps that angle can be quantified from spacecraft orbital (SPICE) data as the flare looks rather similar to the post landing image.
Following another clue, which CIVA camera/image would correspond to the number #2 solar panel being able to see the sunset?
Question for 4th rock, what would be the basis for vertically shifting some of the images in the post #793 cylindrical mosaic? I think the ideal would be to have a cylindrical projection where the "equator" represents the horizon relative to the sides of the spacecraft.
Emily's post has the images on a more consistent level, though I'm unsure what type of polar map projection it is. Perhaps a guess can be made on that one where the "equator" angle would be? It may have been posted what range of elevation angles CIVA can see.
By the way as I mentioned earlier the perihelion sunlight intensity will be about 6 times that at landing.
There are a couple of cues that would help narrowing down the sun direction:
Civa Sky picture: the sun would be in the direction of the broad end of the fanlike flare. The anntenna in that image is clearly lit from that direction aswell.
Civa back lit outcrop: the sun is obviously shining from the side slightly more from the behind. There is also a strong specular reflection in the black carbon fiber rods that make up the landing gear. That is consistent with the light in the sky picture.
Civa rock in sunlight with antenna:
There is a sharp rock there that casts a shadow behind it that is also consistent withthe sun angle in the other two images. If one constrains a plane between the camera and the projection of the tip of the rock and the shadow at infinity you would get a surface on wich you would find the sun. The sun is also at "infinity" So the problem then becomes one dimentional.
I will try to render a model of the landing gear and see where the sun would need to be on that plane to create a specular of that that also matches the small shadows seen on the other feet. (I just happened to build some feet the other day)
There is also a somewhat less reliable feature in one of the other civa images. It looks like there is a caustic reflection of the solar panels shining on the dark rock face. But that could easily be something else. I will build a simple model of the lander body and test that.
One thing that I think should be on every lander is a spherical mirror ball. Even a rather small one attached to a foot would give you a low resolution 360/180 panorama of the scene.
M
jmknapp
Nov 23 2014, 04:53 PM
Here's an ASCII diagram of the CIVA frames from
NAIF:
CODE
Rosetta Lander CIVA-P Frames
--------------------------------------
The CIVA-P camera frames -- ROS_LANDER_CIVA_P_1 .. ROS_LANDER_CIVA_P_7
-- are defined as follows:
- +Z axis is along the camera boresight;
- +X axis is along the camera CCD lines and is nominally parallel to
the lander baseplate;
- +Y axis completes the right-handed frame, is along the camera CCD
columns and points down toward the bottom of the lander;
- the origin of the frame is at the camera focal point.
This diagram illustrates the CIVA-P frames:
.o
o'\\
\\
\\
\\ ^ +Zch1
\\ \
+Ylnd \\ 60 deg \ ^ +Zch2
^ ---------------x-. /
| CH1 \ / 60 deg
| x ---
| CH7 CH2 `
Zch7 <---x +Xlnd | +Zch3 o
o-------> CH3 x---> =====|
Zch6 <---x | o
| CH6 CH4 .
| x ---
| CH5 / \ 60 deg
`-----------------x-' \
// 60 deg / V +Zch4
// /
// V +Zch5
//
//
o.// +Zlnd points out of the page.
`o
+Zch1..5 point 15 degrees into the page.
+Zch6..7 point 25 degrees into the page.
+Ych1..5 point into the page and are tilted
15 degrees away from corresponding +Z axes.
+Ych6..7 point into the page and are tilted
25 degrees away from corresponding +X axes.
+Xch1..7 (not shown) complete the right-handed
frames and are parallel to the lander XY plane.
So 1-5 point 15 degrees down and 6-7 point 25 degrees down. The FOV is 60x60 degrees.
Malmer
Nov 23 2014, 05:01 PM
QUOTE (jmknapp @ Nov 23 2014, 05:53 PM)
Here's an ASCII diagram of the CIVA frames from
NAIF:
So 1-5 point 15 degrees down and 6-7 point 25 degrees down. The FOV is 60x60 degrees.
Great!
Malmer
Nov 23 2014, 05:07 PM
QUOTE (Malmer @ Nov 23 2014, 06:01 PM)
Great!
Do we have a timestamp for the CIVA images?
Malmer
Nov 23 2014, 05:23 PM
From what i hear they have more than one panorama from philae.
So using the info i outlined in the last longer post one would get two sun directions and could then constrain philaes orientation completley.
The cool thing is then to calculate the optimal position in orbit and day time to image the terrain to catch a specular reflection in the solar cells. Should be "easy" compared to the almost superhumanly awesome stuff they have already done...
Malmer
Nov 23 2014, 05:38 PM
Oh, if one does the shadow projection plane trick for multiple shadows one would get a set of planes that would intersect creating a line pointing at the sun... Which was nice.
scalbers
Nov 23 2014, 06:24 PM
Nice info in the past few posts. I'm presently setting up software to make a version of the cylindrical mosaic.
4th rock from the sun
Nov 23 2014, 06:55 PM
QUOTE (scalbers @ Nov 23 2014, 02:56 PM)
Question for 4th rock, what would be the basis for vertically shifting some of the images in the post #793 cylindrical mosaic?
I used the diagram on that post to position each image.
It's a vertical view, and you can see that all cameras overlap at horizon, yes, but not close.
I just followed the matches I saw and scaled the apparent FOV.
So, the camera views that converge closer to the lander are lower.
Those that converge at distance are closer to the horizon.
Of course, I may be completely wrong, but I think that feature overlap / parallax on a surface close to the lander important. We can't just project the images using a set FOV as would be the case of horizon features.
scalbers
Nov 23 2014, 07:02 PM
Thanks for the pointers 4throck. If I could follow up, I found this raw CIVA1 image here. It looks like you have more detail on the dark rock faces than can be seen in this image. Was there another source for a better exposed version? I see a version of this by Spacepoint in post #372.
http://sci.esa.int/rosetta/54964-frame-fro...iva-p-camera-1/Mattias already has a good rotated cylindrical view (post #428) though I'm unable to click on the larger image.
Malmer
Nov 23 2014, 07:48 PM
QUOTE (4th rock from the sun @ Nov 23 2014, 07:55 PM)
I used the diagram on that post to position each image.
It's a vertical view, and you can see that all cameras overlap at horizon, yes, but not close.
I just followed the matches I saw and scaled the apparent FOV. That what, the camera views that converge closer to the lander are lower.
Those that converge at distance are closer to the horizon.
Of course, I may be completely wrong, but I think that feature overlap / paralax on a surface close to the lander important. We can't just project the images using a set FOV as would be the case of horizon features.
There is no "right" way to project these images onto a cylindrical panorama. They are not originating from one point. So unless you have per pixel distances you will introduce errors.
I will treat them as images at infinity until I have better data. (The landers feet hovever are at known distances so they can be registered... )
I keep everything in 3d at this point. That way i can measure things correctly.
4th rock from the sun
Nov 23 2014, 10:02 PM
Here are my source images. Don't remember if they were posted here or not.
The image names are original and indicate what CIVA camera was used.
Click to view attachment Click to view attachmentClick to view attachment Click to view attachmentClick to view attachment
4th rock from the sun
Nov 23 2014, 10:05 PM
The image from cameras 6-7 seems to have been averaged. I can see doubled features there.
Click to view attachmentTrying to assemble the images in 3D space is a good idea. Looking forward to see what will come out of it.
Decepticon
Nov 23 2014, 11:13 PM
I look forward to your image.
I hope the rest of the images get released sometime soon.
Malmer
Nov 24 2014, 12:31 AM
PHILAE SUN VECTOR
I calculated the vector to the sun in the philae spacecraft frame coordinate system using the plane intersection approach I outlined in previous posts.
here is the Philae coordinate system for reference:
CODE
.o
o'\\
\\
\\
\\ ^ +Zch1
\\ \
+Ylnd \\ 60 deg \ ^ +Zch2
^ ---------------x-. /
| CH1 \ / 60 deg
| x ---
| CH7 CH2 `
Zch7 <---x +Xlnd | +Zch3 o
o-------> CH3 x---> =====|
Zch6 <---x | o
| CH6 CH4 .
| x ---
| CH5 / \ 60 deg
`-----------------x-' \
// 60 deg / V +Zch4
// /
// V +Zch5
//
//
o.// +Zlnd points out of the page.
`o
X is forward
Y is to philaes left
Z is up
a normalized vector pointing to the sun:
X 0.0449037
Y 0.68573
Z 0.726469
I am rather confident in this number. it should be good to within 5 degrees. (if you like to trust some guy on the internet)
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please
click here.