Oppy looks like to be stuck in a dune once more!

The front wheels (rover was moving backwards into the dune) look to be closer to solid surface than during the purgatory incident, so hopefully they can back out pretty quickly.

I know this is already being discussed in the other thread, but I think this way the incident is clearer to a casual reader of the forum. Threads of nearly 1000 posts clog the discussion anyway.

http://qt.exploratorium.edu/mars/opportuni...GBP1151R0M1.JPG

http://qt.exploratorium.edu/mars/opportuni...GBP1311R0M1.JPG

inline images changed to links. 2048 x 1024 of images clogs discussions as well - Doug

A whole lot of doom and gloom when we don't know how stuck it is, how much driving took place.

Remember the mini-purgatory back toward the NE rim of Erebus. They were out of it a sol later, Give it 2 sols....THEN maybe there's grounds to be slightly nervous, but even then, we got out of Purgatory, this will be a breeze.

Doug

I agree Doug it's not anywhere near as bad a Purgatory, I think we'll get out fine.

I do think it's much worse than the other event that we got out of straight away though.

---

Was this the other event?

http://qt.exploratorium.edu/mars/opportuni...FSP1235R0M1.JPG

http://qt.exploratorium.edu/mars/opportuni...FSP1335L0M1.JPG

Assuming it was (Edit: it was, thanks Tesh) - the tracking site says there was about 5m of driving to get into that (and the rover made 1-2m) in this case we apparently have 24m of driving. Still, a lot less than at Purgatory (50m+ iirc)

The rear wheels don't seem to be as deep in the muck as they were during Purgatory.
Click to view attachment

Okay Akuo. This significate that it's not desperate?
There is an other thing : the rock layer seems to be less deeper than the first time Oppy was "jailed" by sand. Is the wheel can touch this layer and the rover roll onto?

I must admit that the situation is not clear to me. If you look the tracks, you can see that we moved backward at first then came back to the outcrop and finaly moved backward again... and there is NO sand in the direction we're driving so, we're FREE, aren't we?

I don't think so. Looks like Oppy is plain backwards driving mode for this, and they just zigzagged around that flat piece of outcrop. Oppy executed a turn a few metres back from the place she is now. Then it was straight (backwards driving) into the dune and stopping there for whatever reason - either because of the end of the commanded drive or the slip protection stopping her.

Indeed jamescanvin, it was supposedly 24 m of driving, question are if the slippage detection did cut in or not.
Even though im usually pessimistic about most things, this time im not, to me its looks less severe than Purgatory and more similar to that 'slightly stuck' situation we had some time after that. In short, I view this as one small unforseen delay on the trek towards Victoria and nothing more.

This is what happens when JPLers don't follow Tesh's route :-)

Rover drivers, take your lesson and stick to the blue line :-)))

where's the 24m figure coming from?

Doug

Hopefully no Purgatory-Dune scenario !
It took the JPL-team about 5 weeks to get the MER out of its dug-in position ( Oppy was stuck from sol 446 to sol 484 ) ... fingers crossed

Humm...It looks like everyone is arriving to watch the accident...

I've got hope on seing Oppy out of there in less than a week...Let's vote?...
Just kidding Doug...

That's the distance from the tracking database.

Example:

..sol/832..::71::233::10.5673::-65.9588::-0.0208675
..sol/833..::71::327::12.1891::-89.7695::-2.32683

Last three numbers (I think) are: distance east, distance north & distance 'down' from 7100 in merters. As has been discussed at length in the route map thread these usually tally pretty well with Tesh's route map.

It's also very unusual to see such a large z displacement - I guess what has happened here is the rover knows it was pitched down for the whole drive and so after 24m forward it should have gone down ~2m.

I'll dig up these numbers for Purgatory and Near Purgatory, stand by.

James

Ahh - it may well be that 24m was the expected drive distance, but some in-drive visidom killed it. Those figures sound like the expected position based on predrive imaging.

Doug

I have the impression that those numbers are derived from the wheel turns plus instantaneous rover's attitude but doesn't take into account possible slippage.
It would explain the big difference with the actual movement, which was about 2-3m based on the "wheelmarks" (each 80cm, right?) which can be seen on fhaz images.

I don't think that's the case Doug - there are different numbers for images taken during drives e.g. auto-nav which can't be predicted beforehand. Look at my overlay of Tesh's route map from a few weeks back here. Also there is no way the rover was anticipated to have traveled 2m 'up' - that struck be as unusual when I first looked at the data before the images came down.

Just one remark about the z axis, James.
I think -- if I followed correctly these data on previous sols -- negative values are actually higher altitudes. Again, it would be consistent with climbing a dune for 24m.

Edited: For instance, in the range of site 70xx (from sol 813 to 821) the rover was clearly moving downslope and the "z value" was increasing up to 1.3m on sol 821.

You could be right, I assumed it was down by the right hand rule but, haven't looked into if that fits observations.

EDIT: Ah, OK we actually agree, I misunderstood you, as I wrote the wrong thing in my first post. The z (actually c) vector IS down (-ve up) but I should have written that the rover was pitched UP as you correctly pointed out.

Here is the tracking data for Purgatory and the North Erebus Incident (NEI)

NEI

..sol/602..::62::262::-26.2507::11.2459::-0.143594
..sol/603..::62::308::-29.8538::7.33988::-0.498349

Purgatory

..sol/445..::55::136::2.50168::-73.9078::-0.857053
..sol/446..::55::226::14.2636::-154.324::-5.38479

~80m - Oppy only drove ~40m before getting stuck.

Again notice how Oppy thought she had climbed 5.4m by merrily climbing Purgatory for 40m!

So by my reckoning tosols incident is only about half as bad as Purgatory but 4-5 times worse than the NEI.

We'll see in the next few sols - but I'm not too worried yet. Maybe It'll bring my Victoria arrival date back into play (973)

James

How did it happen? I thought they put lots of drive precautions in place after the purgatory incident?

A few questions :
It seams to me that rear weels didn't get stucked, do you agree on this? Anyway, are the measurements you show been taken on front weels, all weels? Just wondering what's the logic Oppy's got when she feels she's gona be stucked : stop all weels and wait or what else?
We also had pictures of the same sol from Pancam and Navcam. Do you know if Oppy's programmed to shot this kind of pictures even she detect a "big" slipage problem or if the pictures have been taken before been stucked in this case?
Thanks
Climber

A blind drive is still a blind drive. Previously - they would do blind drives "command 6m driving that way" and then autonav "keep going for X metres toward a goal, avoiding obstacles".

Then - visidom could be added into it, "drive for X, then check for slippage, then drive for X and check for slippage again etc"

But the terrain here looked so easy, I wouldn't be suprised if they blind-drived it for 25m. Just bad luck really.

I think the rear wheels are more dug in than the front, as the rover was driving backwards at the time.

I think, looking at the terrain that it was driving into, that actually a bad turn might have occured. I can't imagine they'd blind drive into this stuff..

http://qt.exploratorium.edu/mars/opportuni...GBP1311R0M1.JPG

Perhaps the preceeding turn did not occur as much as expected - I'm sure we'll get an explanation in the near future anyway.



Doug

> It seams to me that rear weels didn't get stucked, do you agree on this?

This is an "enhanced rhaz" image; if the wheels are stucked or not is something we'll know once they put the reverse.

Click to view attachment

> Anyway, are the measurements you show been taken on front weels, all weels?

All wheels I would say.

> Just wondering what's the logic Oppy's got when she feels she's gona be stucked : stop all weels and wait or what else?

Afaik, there is no "felling I'm gonna be stucked" but something like:
- Take a navcam image
- Command an X meters drive (X may be 5, 10, or some meters)
- Take a navcam image and calculate the actual driven distance
- If slippage (actual distance / planned distance) is greater then <param> abort drive, otherwise repeat next cicle.

We also had pictures of the same sol from Pancam and Navcam. Do you know if Oppy's programmed to shot this kind of pictures even she detect a "big" slipage problem or if the pictures have been taken before been stucked in this case?

Those pics are post-drive images, taken once the rover finished the drive normally or not.



This is a partial pancam mosaic taken on sol 828, two moves before the current position, in which I encircled the current location. It really looks easy, doesn't it?
Except for a "fluffy" texture, I wouldn't say anything against going through this path.

Click to view attachment

I think the visidom combo they use is essentially blind-drive-for X metres, then do a short bump drive - compare before and after images of the bump to check what slippage is. If it's low, then blind drive for another X metres. It's a bit like walking a few paces, checking the terrain with a stick, then walking a few paces more. It's complex and convoluted, but just about the only way to do slippage check.

Real time slip-check ability is something that would be of huge benefit to future missions I'm sure ( i.e. Exomars and MSL)

Doug

Speaking of "when should we expect a backwards movement", I would say sol 836 as the earliest assuming almost no further analysis except "put the reverse like on previous times".

This sort of mishap is not unexpected. Although these ripples are 'as old as the hills' by Earth standards, this is a dynamic locale by Mars standards, especially since we are approaching a major topographic feature. This 'change in the sand' happened as we left the Endurance plain and entered the rippled plains, it happened as we approached Erebus, and now that we approach Victoria...

Hindsight is always 20/20, but look at the ripples, they look subtley different here.

This doesn't hurt my arrival time guesstimate a bit: I still say 'eventually'.

--Bill

Pardon me if I missed this info - I am scanning the previous posts - but I notice the IDD is deployed in the original picture.

That wouldn't get deployed automatically after a drive? So someone has to be talking to Oppy today....

Or else they are trying to get shots to categorize the soil they are in.

I love how Oppy chooses a holiday weekend in the US so we won't get any official updates today.

Can't believe this, obviously one of the rover drivers was betting on a late arrival date at Victoria...

Yes it would. It get's pulled out automatically after EVERY drive as a matter of procedure since the trouble they had with it at Olympia.

Doug

Do the MER drive motors have variable speed?
If so, I'd recommend a creep-speed backup, then gently roll west until we're
back on the blue line! If that fails, then gun it backwards.
On the Route Map we seem to be entering a large area of 'different' ripples.
They scare me now.

No, Oppy is deploying the IDD automatically after each and every drive now. The reason for this is the problem with the azimuth motor in the IDD - they want to have the IDD out in the case that the motor breaks during one of these cold Martian winter nights.

What I am not sure of, is that if the IDD would still be deployed after an alert or an error during the execution of commands. I.e. would it be deployed after a slip check alerts the rover?

With it's foot to the floor, MER is creep-speed anyway

Doug

Some sophisticated truck 4x4 has own mechanical slippage blockage. For the MSL, this kind of slippage blockage would be commanded by some kind of electrical/photo sensors.

About the stuck. I have studied the pictures and I am not worried of this. Oppy can be back again by just slidding on the same track backward since the stucked are only when Oppy was trying to get over the bump but luckly it has not already come on the top of bump which is the hardiest part to be unstucked.

Rodolfo


This delay will favor my bet! (I am not thinking badly of she but I took the pragmatic way). I calculated that the travel to VC will have at least of 30 days of any kind of accident or any deviation of the straight line route for any further curiosity that might encounter on the way to VC.

Rodolfo

Yes - a few months ago we were talking about how they could use something similar to the optical sensors on modern computer mice. Somebody ( Sorry - I can't remember who - was it James?) - took the mechanism, added a 35mm lens to it and it worked at quite a range....certainly adequate for WEB-belly->surface measurements.

Doug

[quote name='Tesheiner' date='May 29 2006, 04:35 PM' post='56196']
[i]> It seams to me that rear weels didn't get stucked, do you agree on this?
This is an "enhanced rhaz" image; if the wheels are stucked or not is something we'll know once they put the reverse.[/i]

Thanks so much M.Tesheiner to have taken time to answers my questions . I actually didn't see properly the rhaz image : before your enhanced version I thought rear weels were "on" the sand, not in.
May I come up with another wild idea? I'm (again) thinking of the software that will allow Spirit, in order to follow Dust Devils, to download only pictures when changes occured. Don't you think that this could be used to avoid what is happening to Oppy? I mean, on top of what she does to check that she's not stucked, allow her to take a picture and compare to the previous one for changes : no change = I'm stucked.

> Thanks so much M.Tesheiner to have taken time to answers my questions

M. Tesheiner? I must be getting old.

> Don't you think that this could be used to avoid what is happening to Oppy?

Basically that's visodom (afaik). But another "feature" of this driving mode is that's sloooow; images must be taken, processed and compared with the previous one, and that means CPU power.

[quote name='Tesheiner' date='May 29 2006, 10:59 PM' post='56226']
M. Tesheiner? I must be getting old.

True! That's a by-product of the BBQ pict!

> Don't you think that this could be used to avoid what is happening to Oppy?
Basically that's visodom (afaik). But another "feature" of this driving mode is that's sloooow; images must be taken, processed and compared with the previous one, and that means CPU power.

I thought about the CPU but it's a matter of compromise, and my bet is sol 935 for Oppy arrival at VC . By the way, do we know yet what ship will be used on MSL ? Not sure to have seen this yet.

I think your right about this Doug.

And in fact I think the slippage detection worked in this case, it was just unfortunate that, thinking the terrain was good nearby, Oppy was told to go 24 or so meters before checking. I'm pretty sure the drive was aborted early, at the first visidom measurement.

There are two fs_commanded navcams - at the final point and at [final drive no - 2] (i.e a small distance before) I think this is the slippage detection imagery.

Also note there was no turn in place to line up of better UHF passes and it doesn't look like any 'penultimate' images either, all signs of an aborted drive.

---

The fact that at both Purgatory and here the rover thought it had climbed quite a height by being pitched up for so long while driving makes me wonder why that can't be used for detecting this kind of thing? As Oppy drives over ripples the number of wheel turns plus the rover attitude should map out the ripple profiles in a no slip senario. And we know there arn't any 20x2m wide - 2m+ high ripples around here so stop if you appear to be going over one!

Where are we? Doesn't look like the same place...
Looks to me like the problem was solved with no major problem...:
http://qt.exploratorium.edu/mars/opportuni...VMP2410L7M1.JPG

one of many quite old images coming down the pipe, probably on an overnight UHF pass

Doug

Might as well be from the afternoon pass, since the images have been arriving in the last few hours. Nothing from tosol, so I guess Oppy is staying put while folks at JPL assess the situation...

Aaaaargh!
Will somebody grab me by the shoulders and put on the right tracks?!!?

Nice break.

- http://qt.exploratorium.edu/mars/opportuni...VMP2410L7M1.JPG

That's from Sol 818.

There was no (attempted) driving tosol (834)

--

This is the most significant downlink of the day:

http://qt.exploratorium.edu/mars/opportuni...G9F0006L0M1.JPG

One of the visidom images that aborted the drive. It's a shame the other isn't down yet as we might be able to see how much Oppy was slipping when the abort was called.

Thanks for clarifying that james.

Speaking of slippage detection & driving modes, I remember reading a paper some time ago talking about them.
Here it is.

We're just killing time here to ensure that Oppy arrives at Victoria precisely at Sol 965.

Words from Mr. Squyres:

'We don't think so. The drive was terminated by a slip check, and the rover is not as deeply embedded as it was at Purgatory. We feel confident about extracting it and moving onward. - SS'

They would, Ustrax, but they're afraid you'll hug then to death in gratitude.