QUOTE (Phil Stooke @ Aug 20 2006, 08:04 PM)

... with a grid (based on a map-projected MOC image) at 100 m spacing.
Phil

I counted the grids : we're (about) 6300 m south of Eagle-Endurance. I know you know, but...

Anyone else notice that the Twin Peaks (crater rim near 100 degrees azimuth) has ceased to be visible since moving onto the Victoria ejecta? This blows away all those notions I'd had about using them as a triangulation and calibration reference. I'd assumed this would occur eventually, but figured we'd be much closer to Victoria before it happened.

Phil, I notice that since entering the ejecta terrain there is a noticeable divergence between your map and Tesheiner's map (nice to see all of yours in one place). Presumably the dearth of landmarks is presenting the anticipated challenge to navigation. I'm curious as to what references the two of you are using at this point to derive your positions. I'm assuming that Beagle, Epsilon, and Delta are the most obvious targets, but of course they're all a bit ambiguous from this far away, so what I'm particularly wondering is whether there's enough detail in the MGS images to identify the particular dunes visible around Oppy.

At the moment, both your's and Phil's route maps suggest that Oppy is pretty-near aiming straight for Hawking rock, so it looks like we'll get to know it much better before long. The real trick will be to fix it to a nice unambiguous position on the route map, much as when we first hit Jesse Chisolm (West Hillock). Anyone know whether there is an ongoing campaign to have MGS periodically the rovers with c-proto images? It's been a long time since the last one.

QUOTE (algorimancer @ Aug 20 2006, 11:13 PM)

Anyone know whether there is an ongoing campaign to have MGS periodically the rovers with c-proto images? It's been a long time since the last one.

Sorry I have no clue. I just want to remind that Doug once said that MRO's images will not surpass MGS c-proto's. So, we'd better get some more soon! In another end and IIRC, I red a long time ago that they were expecting MRO's pictures to help explore VC. Some contradictions there.

Nuhhuh.....I think they will MC doesn't think the difference will be much, but I think we'll be suprised, with twice the number of pixels on the ground downtrack, and 6 times as many across track, and a sort of 'colour' into the mix. OK - so it wont resolve 25cm/pixel (but the test images were damned impressive).....but even if it's only 50cm/pix - that's three on the ground for every CPROTO pixel.

Doug

Yes, I've noticed that things (appearances, actually) change quite rapidly on the Victoria ejecta apron. This is a major topographic feature, elevation-wise, but is very wide and smooth so you don't realize how tall it is. Beagle is sinking below the horizon, and look at the detail we're starting to see on the "Epsilon" rim.

I've not been able to guess at "Hawking Rock" on the MOC image, though.

--Bill

After only 2 days of driving since leaving Beagle, we're almost 1/4 of the way from Beagle to Victoria.

"Phil, I notice that since entering the ejecta terrain there is a noticeable divergence between your map and Tesheiner's map"

Don't pay too much attention to exact locations - they will evolve a bit over time as we get exact distances and more looks back. I frequently go back and edit my map. I posted the big one because I've made numerous changes across the etched terrain.

Phil

Best guess at the location of Hawking Rock:

http://www.clarkandersen.com/R1500822_NWVi...mpp_Hawking.PNG
(big file, about 4.5 Megs, uncompressed)

This is a rescaled version of image R1500822 (the map projected version), scaled to 0.5 meters per pixel. It seems in reasonable agreement with the azimuths as reported by the AlgorimancerPG application.

I managed to find a single image showing Hawking from Sol 887, combined with another from Sol 904, yielding about an 80 meter baseline. Performed wide baseline photogrammetry to Hawking to yield the single-pixel hilighted positions. Results of 4 calculations are shown (single colored pixels). I would lean towards the lower two "corrected azimuth" versions (w/corrections referenced to the Twin Peaks yielding +0.33637284 azimuth degrees for Sol 887 and +0.086453315 azimuth degrees for Sol 904 (azimuth correction will be available in APG 3.3, not yet posted). I would allow a bit of leeway in the east-west dimension to allow for error in the image projection, but feel pretty confident about the north-south position (to within a few meters). Bearing in mind that illumination is from the left in this image, several features immediately east and west of the lower two pixel positions may in fact be Hawking, but for the moment I'm leaning towards the dark E-W linear feature immediately to the left (west) of the lower pixel pair, which terminates at the pixels, as Hawking + shadow. In ImageJ, with the image processed to apply a shadow illuminated from the Northwest, that feature stands out nicely.

It would be interesting to look at azimuths to Hawking from more recent Oppy positions, but for now there is some fuzziness in those positions, and no good horizon reference to calibrate the azimuth.

After reading the planetary.org report about MER, I have found an phrase which Arvidson is saying that the MER will climb onto the annulus of Victoria and drive across the northwest portion of Victoria. That place is close to Beacon at near rim.

The extracted article.

While the discussions and debates about which direction to go and what to examine up close are ongoing, there is a general plan, Arvidson said. "Beagle crater features ejecta rays [areas of ejecta that look like rays in the orbital images], so we'll explore some of those first, then we'll climb onto the ejecta apron or annulus of Victoria and drive across toward the northwest portion of Victoria. We want to characterize the broad, dark apron [the Annulus] that extends out around the rim of Victoria, because it's very peculiar, very homogenous when seen from orbit, and we're not exactly sure what it is." Nor do they know what it will take to cross it.

Rodolfo

Just a side note, Irfanview can compress (losslessly) PNG files quite well. Get the Plugins pack and you can use PNGOut, which can really compress PNG files. I ran your file there through Irfanview with PNGOut, and it's down to 2,684,928 bytes.

And now back to your regularly scheduled thread....

For what purpose? We know where the rover is, and it's hard to improve on s11-471 for route planning.

Not sure what people are expecting. I doubt that the rovers will be resolved as unambiguous objects. I think the images will be improved somewhat over CPROTOs. Having more coverage will be an improvement, of course. As I have said before, the real improvement is in SNR, but it's harder to judge what improved SNR will look like. There wasn't much overlap between MOC images and the HiRISE test images, so we'll have to wait some more to see.

I'm a little late, Phil, but your composite route map is quite nice. And I can remember wondering if we'd reach Endurance Crater!

Given the absolute lack of outcrops here on the apron it's almost impossible to (at least to me) match ground vs. orbital images like while driving on the etched terrain.
For sol 912 drive there were some rocks visible on both pre-drive (sol 909 pancams) and post-drive (sol 912 navcams) images so I just calculated the driving distance using APG. For the driving heading, as "Delta" was visible on the pancams, I used something like "X degrees left of Delta".
For sol 914 drive I didn't calculate anything; just used the rover's position info obtained from the pancam data tracking. I plan to continue using it while we continue on site-id 75xx, if you know what I mean.

As I said before this approach accumulates errors so from time to time a sort of correction/adjustment should be made using a different method. Since it looks like they are heading the rover to that ramp at the end of my "hypotetical route", I'll wait until we are very near to "Epsilon" and then will try to pinpoint the rover's position using that crater as a reference. If I succeed, I'll then update previous positions (since sol 912) to match with this last measurement.