Given the current evidence, I think the current pessimism is unfounded. There have been previous events that looked like game over, and were dealt with - think about Spirit's early flash problems.

Some evidence - in Steve Squyres' latest update says the following:

"After yestersol's drive, all six wheels we're dug pretty deeply into a large ripple. We've gotten dug in before and gotten out just fine, so this isn't cause for immediate concern. But we're likely to be here a little while, taking our time to get our wheels back on top of the soil again, and also taking some time to figure out what's different about this soil and how to keep this from happening again as we continue to work our way toward Erebus."

He doesn't seem unduly concerned. Unless we hear otherwise from an official release, or one of our valuable insiders, then I don't think we should be either.

Chris

Looking at this carefully, what is curious is that at least one of the wheels didn't sink in, which suggests that the surface is not very light and fluffy. Look at the right hand side of the track. I wonder if the different dynamics due to the stuck steering actuator have caused the wheels to break through the crust on the dunes.

For those who ski - think about skiing on breakable crust - if you're very gentle, you can stay on top. Too much force, and you break through.

Chris

Another interpretation is that when the front and middle wheels went over this spot, they were on top of the surface. But by the time the rear wheels went over this spot, the front wheels were in the soft stuff and there was resistance to forward progress from the deepening flotation of the front wheels. So the rear wheels began trenching as well.

[By "front" and "rear" wheels I mean the leading and trailing wheels.]

Good idea - sounds plausible.

Or a combination of yours and Chris's explanation. What we do witness here is a sudden change in material, don't we ?

My first reaction was: get back! But now Oppy is in there, they might as well take the opportunity ( ) to investigate the soil on composition (and then get back!). Maybe the lighter coloration doesn't have to do so much with outcrops (like orbital imagery suggests), but with a lighter toned (higher salt content?) material. Salt is lower in density than rock, which could be an explanation for the different mechanical properties Oppy is facing now.

Maybe there aren't any outcrops at all between here and Victoria! Maybe the lighter colour isn't a rocky rim of Erebus cropping out, but completely eroded and fragmented "mothermaterial" of the ancient crater, presenting itself just by a lighter soil.....

In that case, reaching Victoria will only be possible by avoiding the light stuff and find a route on the darker soil....very interesting !

Hmmm... when you look at the highest-resolution MOC imagery, it's true that you can't tell for certain if the "outcrop" is really outcrop or if it's just piles of lighter-colored sand. But if you zoom out a bit and look at a slightly bigger picture, the "outcrop" organizes itself (at least to my eye) into impact-related landforms -- ancient crater rims and "splash pattern" ejecta texturing. That argues for the areas that appear to be outcrop to actually reflect ridges in the local landform.

As has been discussed elsewhere, the blueberries seem to provide a natural cover to prevent further erosion of the evaporite layer out on the plains... and as we advance into the lighter-colored soils and dunes, the blueberries seem to be diminishing in abundance. Maybe we're getting into an area of evaporite that developed a lot fewer concretions? That would mean the evaporite would erode primarily, with no cover of concretions to stop the erosion process.

A good lesson we can learn from both terrestrial and lunar geology is that a vastly large majority of the materials at any given location on the surface of a planet originated(*) within a kilometer or two of that location. I think that argues for the lighter soils and dunes to be the eroded dust from the evaporite layer. I bet the IDD work will show these soils and sands to be very similar in composition to the evaporite layers.

-the other Doug

(*) -- when I say "originate," I'm talking relative to the local landforms. Obviously, large-scale cratering tossed a lot of the original crust around, especially in the ancient cratered southern highlands. But ever since the last of the great impacts and the last of the great floods, I'd bet that nothing much has moved more than a few kilometers in any given direction, with the exception of a very thin layer of dust. DVD

Either that, or else make the things much much wider. Or make a walking rover with snowshoes.

I think it is rather unlikely that this is "end-o-mission". They have actually been deeper in the dirt than this. When opportunity tried to exit eagle crater, they dug in very deep. It was in the middle of a blind drive and the rover got itself out. They are obviously on firm soil or it would be sitting on its belly by now. In fact, it is pretty clear from the images that they were making forward progress right up to the end of the drive.

Wash your hands and take a deep breath. They are going to get out of this. Having said that, this is probably the order of business in "the etched terrain". Which doesn't seem so etched to me.

ed

I'm not so sure that this stuff originated here. I'd bet this is the "global dust". The difference between meridiani and other landing sites (like gusev) is that meridiani is not contributing much to the global dust - it is just a collector. So it tends to be cleaner there. The dust that has been falling from the sky for the last several billion years has to go someplace. There must be something about the region of the etched terrain that tends to accumulate the dust.

ed

Steve's April 27th update still sounds optimistic, but he is making it clear that they are being very careful to "do no harm."

http://athena.cornell.edu/news/mubss/

As he says...
"A note to all you Opportunity fans: Get used to the current scenery, because we're going to be here awhile. "

Something about his tone in this update makes me think he reads this forum. He's using very calm and reassuring language that seems to be directed at a level of worry that I don't think you'll find anywhere else (in public) except here

As someone experienced with having a vehicle stuck in deep snow, I know that overdoing maneuvers early on can get you buried worse. Take your time, Oppy, we'll still be here watching...

Cool..at least we now have some idea of what strategy they intend to pursue.

Strangely none of the space news sites have picked up on the problem yet.

I had the same feeling reading that update. "No apparent progress in the images doesn't mean anything other than that we're being very careful to do our jobs right." Sounds like he is specifically talking to people who follow the images uploaded to the jpl and exploratorium site to keep up with what the rovers are up to, like the people that post to this forum!

Intesting mix of colors and textures of soil in that image. I wonder whats holding the plates of soil together.

He probably is - we're the MER fanclub afterall, and everyone writes newsletters to their fanclub

Doug

Hey Steve, if you're listening, give us a sign.

ed

I can well understand why he wouldnt actively participate here. He's a hellishly busy man with a very busy mailbox, and he doesnt need us lot going nuts at him

Doug

Stitching these Pancam track images, there are many interesting details:

In the inset images I magnified two portions in order to have approximatively the same scale (at least in vertical direction, based on the tracks size) and there is a clear confirm of decreased blue berries concentration in the "bad" terrain; maybe also average size is different (slightly larger)... any hypothesis on origin of this sudden change?

Doug, you're right, for example we would always like to know (bother) a lot trivia - he would get a very busy job here

The other track of "digging" (with another incidence of light):

(480KB) http://www.greuti.ch/oppy/oppy_sol448_L257.jpg



Uups, Marco, I think we need slowly a work plan

How much force does the IDD apply when using the RAT? For it to work it has to push against the target..... Im sure I remember them abandoning a RAT attempt back in Endurance Crater with the rock Wopmay because the IDD pushing against the rock caused the rover to slip (move) ?

Best case is somewhere around 80 newtons, but the value depends on where you are in the workspace, including the direction you are pushing.

The capability is mainly limited by the actuator torques, which you can find in Table 1 of this paper. So reaching out and then pushing sideways is more stressful on the arm than reaching down and pushing down.

If the mobility system can't get itself out, the arm is not going to help appreciably.

Looking at the images, seems to me that the soil has stuck between treads on the wheel, causing loss of traction. The wheels are now almost totally smooth. It will be interesting to see how they can get out of this.

Also, if the area was rocky, the IDD *may* be of little help, but in this case I doubt it's of much use - it would be like sticking your finger into a loose sand.

Based on today's Steve Squyres' update, seems very likely that he really did hear us... Many thanks for keeping us updated!

Smooth wheels:

Maybe they need to change the "6 wheels on Mars" to "6 wheels in Mars."

What I find very interesting is the way that the treads seem to indicate that the leading wheels (the rear ones, since Oppy was driving backwards) got stuck or stopped before the trailing wheels dug in. If I may be so bold to copy from Tman and dilo's fine work:

Notice how the first wheels tracks (the ones under the others) appear to make it over this section without digging in. The pacing of the treads appears regular, and there doesn't seem to be any mound of soil built up. Yet the last wheel track on top seems to "pile up," which would imply that the wheel was spinning at the same rate but it's progress forward was hampered somehow. The tread marks get closer and closer. It's almost as if it was pushing the dead leading (rear) wheels like a plow until it dug in deep.

Could a software malfunction maybe have caused this if it made the motor speed of the wheels to vary independently? Or worse in opposite directions? Maybe it really was a 6 wheel digging routiing!

I think the simplest explanation to that is that by the time the trailing wheels rolled over the same area, the leading wheels where already higher up on the dune crest starting to have problems with traction.

What this actually shows is where the problem started. Take the center of those images where the upper track becomes messy, and then add the distance between the outer wheels towards the direction Oppy was going.

Edit: doh, right, lyford, I just read your post again and that's exactly what you said I need to go to bed now...

(oh heck, bed can wait)

What I find interesting are those rectangular compressed clumps of soil in the tracks. Looks to me that this soil is somehow different -- the compressed clumps started sticking between the wheel treads, first falling out as the wheel turned, leaving the clumps laying there on top of the churned up soil. Then all of a sudden, the clumps no longer fell out of the wheel treads sticking completely to the wheels instead, causing a severe loss of traction.

Earlier, I mentioned that it appeared Opportunity could be high-centered, meaning that it's underbelly is supporting some or all of it's weight, because it rests on the soil. Others have made similar comments.

As someone who has gotten various vehicles stuck in deep snow, mud, and sand, I can apreciate the situation. Is anyone familiar enough with the rover's design and location to comment on that possibility? The cad drawings I have suggest that could be possible in this dug-in, dune-straddling condition.

I have been looking at anaglyphs of the dunes in this area, and from prior sols. (Thanks, MMB) Some of the dunes around here are really quite large compared to those from earlier in the mission. (I'm tempted to say huge.) Some of them were easily navigated by Opportunity, while others were deeply carved by the wheels. The hazcam anaglyphs from sol 439 were quite revealing in 3D compared to just looking at the 2D images. That's where they were doing the initial turning tests of that faulty steering, iirc.

Finally, one thing that hasn't been discussed is a navigation software problem. We heard that there were software issues. Could the software have caused the vehicle to plough itself into this predicament?

I don't think so, looking at the hazcam images, the right side wheels (left side from the point of view of the last drive) are only just over the top of the dune (the trailing wheel (the broken one) is infact right on the top). Given that none of the wheels are completely buried and that from my reading of the CAD file there is about 3cm from the top of the wheels to the underbelly I don't see any way that the rover could be touching the ground in it's current position.

James

Or it could be the other way around that the rectangular compressed clumps only stopped coming out once the wheels were spinning with no traction and hence any force from the adhesion of the clumps to the ground was tangential to the wheel rather than normal to it.

What I'm not sure about is, was Opportunity still making progress untill it tried to turn in place? The front wheels don't look any further dug in than the back ones, suggesting that it was not digging itself in further and it had just gone over the dune crest so the going should have been getting easier.

James

Good questions and observations. I've been noticing that the physical character of the soil has been changing on this drive southward. Whether this change has been due to a difference in the source material (the "bedrock") or a change in the transport of the weathered/eroded material (fewer blueberries to make a desert pavement) or a combination of both is the question. Clearly, the current difficulty is caused by the surface's inability to provide adequate load bearing capacity. In a way, this should have been predictable: Oppy got "stuck" in Endurance as she approached the sandier dunes material in the bottom of Endurance just as she is currently stuck as she approaches the sandier material near the etched terrain.

Haste makes waste: we've been hell-bent to make record drives towards Erebus/Victoria when we should have been stopping to smell the roses and examine the changing soil conditions.

Hindsight is wonderfully 20-20, no? We'll be OK...

--Bill

I am not sure that the rover has "grounded" it's belly:

(From "MER Manual")

Since Oppy was driving up the hill backwards, this pic would show how the rocker bogie (Rocker-Bowie?) would protect the underbelly. Even if she had completely crested, the middle wheels would have to be completely buried and then some before Oppy would run aground.

Does anyone have info on how high the crests are at this point?

"Arrgh, Matey, there be dangerous sand bars ahead!"

We may have been eager to get to Erebrus and Victoria, but I think the Rover drivers were a bit more cautious. On sol 443 they did stop to make some images of the soil. And on sol 439 the drove back and forth across the crest of a larger ripple with no apparent problem.

The caking of the soil in the wheel treads is interesting. If you look closely at the wheel on the left side of the image in Pando's post it looks as if some of the lower treads are free of the sediment. The right wheel seems a little bit higher up on the dune. It's treads appear to be packed pretty tight.

Perhaps just a teeny bit of wirebrushing the wheels with the RAT might be helpful? Incidentally the best way of getting unstuck in both snow and sand if no external help is available is "rocking yourself loose", i e try to drive in the direction you want, then unclutch and let the vehicle slip back. It will go a little way past the equilibrum point and then you cut in traction and go with the rebound. If you repeat the cycle a number of times and time your bursts of power carefully the oscillation will gradually build until you tear yourself loose. It actually works better in sand than snow since you never lose traction completely in sand.
It's hard on the clutch though and I wouldn't like to write software that can do it since it requires quite good timing to work.

tty

What about using the brush/IDD to "blow" some of the sand off the rover's wheels?

New press release has just been posted to JPL:

http://marsrovers.jpl.nasa.gov/mission/sta...tml#opportunity

Well, it says that this is only a 30 cm/12 inch high dune! We can't be on too much of a tilt! That really shouldn't be a deal killer - so there has to be some other factor involved.

But I am not sure how to take the last bit:

Rather has an air of finality, don't you think? I do hope it is unintentional.

They always write it that way?

You're right, they often end with the odometer reading. I guess I am projecting my apprehension!

Something like this?

Since we have several people here with an inside track to JPL, maybe someone can answer this question.

What do they do in the test bed to compensate for the substantial physical variations compared to the Martian environment. Obviously the test rover carries a lower mass to compensate for the lower gravity. But what about the issues of soil mechanics, especially in the situation that Opportunity is in now?

My guess is that in 1/3 gravity with 6 mb atmospheric pressure, the fine particles behave in a pretty funky way. Have they been making these comparisons during normal operations (i.e. micro mass-wasting behavior during trenching) in anticipation of situations like this?

These may seem like trivial concerns, but with Opportuntiy now buried up to her axles in fine silt and salt I suspect those variables are starting to enter into the testbed simulations.

Yes, perhaps wheels like that. With that wheel design though, I wonder if they should be articulated, so that if it encounters soft dirt, the wheels could tilt sideways, putting the larger area on the ground.




I've wondered about simple electrostatic attraction too. If you've ever dealt with flour in the winter too, it can stick together quite well, with minimal pressure. These are fairly heavy rovers, pressing down on a relatively small area. That would cake the fine sand pretty nicely into the treads, though I don't know how much this would impede traction, as with each rotation, the treads hit the duracrust again, and any large pebbles, like the blueberries. This would tend to dislodge some of the packed in sand, if only for a moment.

I reckon (see elswhere!) it's clods that have done the blag; the things have stuck, then stuckstuck, then STUCK and thereafter it's gonna be quite a problem!

The dust/soil sticking to the tread of the wheels isn't anything new.....it was seen on the wheels of Opportunity when trying to drive upto Burn's Cliff - in that instance the left wheel was almost completely buried. It was also seen very recently with Spirit when trying to climb Husband Hill

Hypothesis for the change in blueberry concentration? They're heavier than the light particles of sand, and thus aren't blown onto the dune crests as easily.
Any of them that would get blown to the crest might roll back down to either side of the dune with lighter winds.


I find the false-color picture linked by Tman interesting too. It's blue for the most part, but look along the right side of the wheel track - reddish material is visible. There's a lot of it right near the first dune crest, and then some of it dumps out further along. There are in particular two largish lumps of it, which appear to be at regular intervals, compared to the square designs from the wheels. So the material seems like it was stuck to a certain portion of the wheel, perhaps only falling out when the dirt clod was at the top of the inside of the wheel.


A thought just came up, concerning the cohesiveness of the dirt. Awhile ago, one of the rovers did an inadvertent sample acquisition using one of the IDD tools; I think the Mossbauer spectrometer did it. The rover stuck the instrument in the sand, and then went to do an integration on the one front magnet, but some of the sand had stuck to the smooth metal, and was transported to the magnet. Static, or perhaps even (though unlikely) moisture temporarily pressed out of the sand.

im not sure you could write software to do this. this technique doesnt just require timing, it requires kinesthetic feedback in real time. (meaning you alter the timing based on what you feel happening)
guess it would be worth a try when the list of options get short.

this is a great forum. this is the ONLY public forum i have ever seen where folks dont argue just to be annoying. very refreshing.

kholmar

You could write software to do it, given the correct sensor suite, which the MERs don't really have. For example, we have robots inserting valves into engine blocks with very tight fit tolerances. The robot's tooling has a multi-axis force/torque sensor that feeds back into the motion software, which adjusts the insertion vector in real time. Special case stuff, certainly, but it can be done. However, I'd assume that a rock-yourself-out system would be hard to mass-justify for launch.

You're right that it would be difficult to get the MERs to do this. I think they'd have to look at the inertial guidance sensors and make adjustments in a very tight loop. It'd be a challenge...

I appreciate the civilized tone of these forums too - before I heard about this place, all I had were the JPL pages and the very few crumbs of value to be found on Usenet.

I like it here a lot

Same here; Doug's Place is a sane haven in the rough seas "out there".

--Bill

Yeah, I like it here a lot, too. A very welcome change from the signal-to-noise ratio of the Usenet newsgroups. I hardly ever visit the newsgroups anymore, and I think my blood pressure is the better for it...

Back to topic, the other real issue with either MER rocking itself out of a sand trap is that they don't have the horsepower to get the kind of momentum going that would be needed for such a maneuver. Even at a charitable estimate, the MERs creep along the surface more than they motor along. At the speed we can get the wheels moving, you can't push forward and then let gravity pull you back farther than the distance you pushed forward, which is how the rocking maneuver has to work.

However, unlike the dilemma all of us face when we get a car stuck in the snow, Oppy has an advantage -- not only does it have four-wheel-drive, it has *six*-wheel-drive. And four of the six wheels are independently steerable. That really does open up more possibilities than most of us are familiar with from terrestrial driving experience.

-the other Doug

cough-three-cough

Doug

Um, yeah... well, actually, the "problem wheel" *can* be turned (according to the initial reports I read), just not through its entire designed range of motion. Though I don't think I'd risk turning it much from the closest they can get it to straight-forward, that seven-degree "toe-in" it's set on now. I'd be afraid it would get stuck in a far worse position.

-the other Doug

Wheels in the sky aren't turning yet....

Here is a neat video from the Canadian Discovery channel, which The Supreme Canuck shared over on Bad Astronomy....

Seems like the best bet is to just back out, so I guess the first thing we would see is the wheels straightening out. Which hasn't happened yet.....