The auto nav software does not work that way, so it cannot be commanded to do that. It could be reprogrammed to do that, but it would require a development effort and a sw upload. That would take tons of time.

Short answer is - no, it can't.

I'm down with that, let's be careful rather than fast.

I think fast is what got them in this mess to start with. If they want to do a long, blind drive, break it into small chunks with a short, bump, auto nav drive in between. That way if it gets stuck, the auto nav sequence will go out on fault and stop any further motion. I'm sure that there will be no more 90 meter blind drives in the future.

Ed

Not in the current terrain, of course not. But when we get out of this, we'll be traveling out of this terrain into different stuff -- and if they decide to retreat back to "safer" ground and approach Victoria from a different angle, over the darker ground with 5cm drifts as oposed to the 70cm+ dunes in the etched terrain, we could easily get back to 100m blind drives again.

-the other Doug

Played around with excel and made a graph showing meters driven summarized vs. cm gained summarized.

It shows the very linear motion very well. Maybe with a slightly better tendency at the end. This only goes to SOL 475 though.

Actually it shows a slightly worse tendency at the end. Embarrasing that I can't even read my own graph

Perhaps. It's also likely that they will look for exposed outcrop material where the sand is not deep, and try to navigate the dunes by staying between the crests.

I think the rover tracks show that most of the time the rover stays on top of the sand, but if it starts digging in and continues on for a while it will be pretty hard to get out of it, especially if the surrounding area is flat with deep sand. It can practically rototill for miles without ever getting on top of the sand surface.

I really enjoy having some cold, hard data to play with. In particular, I like to plot data in ways that help me make sense of it. Dot.dk already posted the data in a really interesting plot of cumulative meters driven versus cumulative centimeters advanced.

I'd like to post two other ways to look at the data. This first plot is a simple scatter plot of the centimeters of progress made in each drive attempt versus meters driven, rather than the cumulative values. (Please note that my graph plots centimeters advanced on the Y axis and meters driven on the X axis. I realize this is the reverse of the convention dot.dk used, but I had already made my graphs and I am too tired to change that right now.)

This kind of graph allows one to calculate the slope of the best fit line to all the data points. The slope of that line represents the average advancement efficiency ratio of all drive atempts so far, and it is essentially equal to the 0.5% that has been published. The R-squared value on the plot indicates the data are not widely scattered.


The second graph is even more interesting, I think. It displays that ratio for each Sol on which there was a drive attempt. It's not a pretty picture. Progress was best when Opportunity initially tried to back out, and it has sytematically gotten worse all the way through to sol 475. You can see the 0.5% plateau between sols 468 and 472, but it drops off after the no-drive problem they had on sol 473.

Sorry for the lengthy discussion, but this looked significant to me.

**Edited to embed the actual graphs into this post. These are the same images that are in the links.

Very interesting plots, Dot.dk/CosmicRocker. Hope decreasing trend in last one will not continue!
The "no-drive problem they had on sol 473" should be represented with a 0cm effective advancement or simply no attempt to move?
Thanks.

I really hope that decreasing trend in the ratio does not continue, since it means it has been getting harder and harder for Opportunity to make progress. It is encouraging that we continue to see decent progress, but that is only being accomplished with more rotations of the wheels. I've got to think that somewhere along the way we will see it break upward. That will be the sign that Oppy is breaking free.

I didn't enter 0,0 for the no-drive day to keep it from skewing the statistics for the average ratio calculation in the first plot. There is actually a 0,0 point inherent in that calculation, since the best fit line is constrained to pass through the origin. Also, if I used 0,0 for the sol 473 ratio in the second plot, it would cause the forbidden divide-by-zero error.

487,488, and 489 still available. Anyone want to claim them?

I'll take 487.

I'll claim 489, if at all possible.. I already claimed 470, but admittedly that was not the proper sol. EXCEPT that that was the day it went over that one really bad patch, you know, the one that looked like all the others, but uhh, ehh.

Yeah. I'll take 489, if it's legal to do so.

So, who picked Sol 484?

"So, who picked Sol 484? biggrin.gif"

I did! I just downloaded the images with MMB 1.1 - looks like Oppy's free! Free!

Bubbinski

Wooooohooooooo, this is SO GREAT!!



I'm so happy now!! Go Oppy Go Go!!!

Yep, the front hazcam in the other thread shows that the wheels are on firm ground and headed up and over the dune, and the rear hazcam (the previous message in this thread) shows good traction on the rear wheels, although the whell on the right side of that image seems to have more slippage.

Good deal. I knew she'd get out eventually.

--Bill

Congratulation! About my (bloody close-by) bet sol 483 I found a nice verbalisation in an English dictionary: "A miss is as good as a mile" - right?

Can almost see the cartoon version of this already. Two small martians sitting on the solar panels of Opportunity one saying to the other: 'Let go of the wheel and move over, from now on I will drive!'