Click to view attachment
I was just looking over the tau images on Sol-1797 on Opportunity, and I found somthing pretty odd. At first look I thought maybe it is just a cosmic ray hit or just random noise, but when put into into a gif, it looks to be somthing different. Any ideas?

Thanks for the answer Paolo! We will all hold good thoughts that the RF wheel current returns to normal.

Are they hot pixels? There are two that move in formation. Or I should say, the sun is moving, and the pixels are stationary in formation...

I'm sort of now thinking that .

Um, yeah -- I mean, it's either hot pixels or you've caught the Barsoom Airlines flight pattern out of Helium, all the commuter aircraft bound for the great thoat roundup...

-the other Doug

A couple of new 3D images on my ROAD TO ENDEAVOUR blog if anyone wants a look...

http://roadtoendeavour.wordpress.com/2009/...dows-and-tracks

Today's (sol 1802) imaging plan has only a tau measurement.

This was a long weekend (president's day holiday). The Sol 1800 currents were still high, today it will be one more diagnostic test I believe.

Paolo

Oops, I wasn't aware of the holidays.
With the low activity today (in terms of MB) I hope those images looking toward Endeavour get a chance on the downlink queue.

Well, if we are heading toward a wheel lockup, I think it's time to tell Oppy, no more bathroom breaks till we reach the parking lot terrain.

Paolo, any plans to experiment with that wheel held locked just to see how well she moves through the ripples? If it is not too big an impediment to travel, could locking up the wheel actually extend its lifespan for situations where all 6 wheels are really needed?

That's why she ended up with one wheel locked toed-in, she was trying to hold it ;-)



I have no idea (and probably the rest of the team as well) whether this is a symptom of actuator wear. W know that backwar driving reduces the current a little, so we might decide to drive backwars for a while. That is mutually exclusive with Autonav/D* tho.

Regarding extending the lifespan of the wheel y not using it: when I moved to the US a long time ago I bright with me a PAL VCR. It was well used when I moved, and it arrived in perfect shape. I used it every once in a while to watch tome PAL tapes. Then I decided to use it sparingly. In about three months of being unused the drum stopped spinning freely (due to lubricant breakdown).

So if they will ask my vote, I will vote to continue using the actuator for as long as we can. There's always time to test whether we can cross ripples when the wheel stops working :-(

Paolo

How much is influence of the sand is driving through? Quite recently she moved from bedrock to more sandy areas, and it seems to me the front wheels take the brunt from this they 'dig the track' that the other wheels are following in. A bit further on she will be back on bedrock patches, so hopefully this will help.

Regards,

Geert.

That is my personal take (that the higher wheel currents are due to the soil type), but I have to be honest and say that the increase is quite a bit larger than we have experienced so far. We did see higher wheel currents on the RF prior to VC, on the north side of VC and in the "parking lot" south of VC, but the difference between the RF wheel drive current and all other wheels was small compared to what I see now. So at this point my *hope* is that this is due to the soil.

Paolo

I think we all share this hope with you, lots of positive waves must help

Looking at the rear hazam images on sol 1800 and 1797 it looks a bit like the right wheel was digging deeper into the soil then the left wheels. The driving she is doing now is different from driving on 'parking lot' soil, when she is driving south in between the dunes there are often situations where one side is more or less 'on' (or 'in') a ripple while the other side is in a through, it stands to reason that in such a situation drive currents will be higher on the side which has to dig deeper through the soil. On previous 'parking lot' trips all wheels were on equal footage and that would make a lot of difference.

On Spirit I'm always amazed that the 'stuck' wheel is really 'stuck', it is not turning at all, normally with a burnt electro motor you would expect that you could still turn it (even though it would show some resistance), but it looks like the wheel is full on the break, everything would be a lot easier if it would turn at least a bit...

Regards,

Geert.

Probably I'm missing something but if I remember correctly then wasn't Oppy used to regularly alternate between forward and backward driving for better "wear distribution" anyway ?
Or has it something to do with the stuck IDD or the new flight software that could make regular backward-driving more difficult ... ?

If I understood correctly there is no problem with driving backwards; the limitation is driving backwards in Autonav mode.

In another topic, those pancam shots taken during sol 1797 and looking on the general direction of Endeavor were finally downlinked on the last session. Look for the images labelled like "1PxxxxESF97PRP2409" here: http://nasa.exploratorium.edu/mars/opportu...cam/2009-02-17/.

01797::p2409::19::16::16::0::0::1::33::pancam_endeavor_look_R12

New rover poem here if anyone wants a look...

http://astropoetry.wordpress.com/2009/02/1...fter-five-years

Makes me wonder what's more dangerous; backwards driving with no autonav or forwards driving with the risk of losing the use of the RF wheel.

That's because of the gears between the motor and the wheel. Too much friction!

Paolo

Autonav cannot be sequenced by driving backwards since we have to use the NAVCAMs to look at the obstacles (there isn't enough texture for the HAZCAMs) and the LGA is in the field of view. Yes we used to alternate back and forward driving but if we want to make some serious progress we need to use Autonav. That would force us to drive ackwars blind, do a 180deg turn and then drive with Autonav. The issue is the 180 deg turn. It is a large turn and it would put quite a bit of stress on the RF wheel. This is why so far we drove forwards. Pre Conjunction we did some drives with backwards blind, turn 180, Autonav. It might be that we have to resume that strategy.

Paolo

There is no more danger in driving backwards blind. In fact under certain conditions it is safer to drive backwards!

Paolo

Umm.... does anyone see anything that might be Endeavour?

I couldn't see Endeavour in those pancams. We have seen parts of the rim recently in navcam, though. The most recent view I've noticed was sol 1794:
http://marsrovers.jpl.nasa.gov/gallery/all...A0P1615R0M1.JPG
Anyone know how the azimuths of the sol 1797 pancams compare with this navcam view?

Hum...maybe I'm seing things but isn't Endeavour present in this image?
http://qt.exploratorium.edu/mars/opportuni...00P2404L2M3.JPG

The double-bump feature in the Navcam is around azimuth 91 or 92. That's about one-third of the way from the left in the first Pancam image (1p287716558esf97prp2409r2m1).

I don't see anything.

looking at Tesheiner's latest route map I can't avoid to suggest a song to go along with it...good mood!
Here's my own updated map...
http://i16.photobucket.com/albums/b14/ustr...01_leg2_01f.jpg

Here's an interesting note included in the latest status report: "The plan for the near term is to drive backward to see if that improves performance of the right-front wheel, although backward driving will limit the distance traveled each sol."

Well, you are not seeing 160m drives for some time but we have done quite long blind drives. I believe I have done a 120-140m blind once.

Paolo

That's no problem: Even 100 meters/sol is incredibly fast progress if it can be regularly sustained over a substantial amount of time (in particular when taking fully advantage of non-restricted sols) ...
Looking forward to a streak of several weeks of continuous 50-100 meters/sol drives ... which would already bring us in reach of the "parking lot" terrain

I presume that in blind drives the usual "emergency stops" like maximum tilt and wheel slippage are still operating?

Attitude information (e.g. tilt) is obtained from the IMU but wheel slippage needs visual/navcam data. Does visodom work with navcam images pointing opposite to the driving direction?

Something just occurred to me (probably occurred to others, but I am too lazy to search the forums).

Given that Oppy has already lost the ability to steer one its wheels (fortunately while it was pointing almost straight ahead!), why does Oppy still perform turns using the steering motors? If one of those motors were to die while a wheel was turned at a high angle it could be curtains for future rover control.

Why not instead point all the wheels forward and turn using a 'skid-steer' approach (ie like a bulldozer or 'bobcat'). Yes the turns would be a bit gross and wide, but who cares, what are you going to bump into out here? This includes the end of drive turns for downlink orienting, etc.

Paolo, is there a reason skid steering is not acceptable or why avoiding skid steering is worth risking a turning motor locking up while turned?

My guess is you would just be trading wear and tear on the motor for perhaps greater wear and tear on other parts. Also the loss of steering control in a region full of sand traps would not be good. Also, energy is at a premium up there and that sounds like an energy intensive way to drive. I bet there are a lot of other good reasons.

That is correct. Tilt, pitch, slip values are all monitored.

Paolo

It actually works better when you are driving backwards. In meridiani the featured that VO uses are mostly the tracks. Driving backwards they are just in front of the vehicle. When we drive forwards we have the LGA in the way so we have to point the NAVCAM slightly off from the straight back and therefore see the tracks at an angle and larger distance. It usually works but sometimes it does not.

Paolo

Correct. In addition, and this is more for Gusev type of terrain, turning while the wheels are straight, 1/2 of the act as a scoop and is a rock falls into one, as it happened on MER-A 318 (I think), it is a potential wheel hazard. If the rock (we call them potatoes) gets wedged between the wheel and the actuator assembly it might prove difficult to get it out.


Paolo

I have to admit, it's fun to see that on JPL parking lot...

Paolo,

Assuming you drove the equivalent (terrain wise and energy wise) to a drive like Sol 1797 - which achieved 111 meters - how far do you think the rover would have travelled with a jammed (please not!!) wheel? That might help all of us here to extrapolate to Oppy's likely future progress if the worst did come to the worst.

All else being equal, do you think we could still get to Endeavour?


Hugh

Excelent question to which I have no answer. We gained lots of experience with Spirit in this category and one thing we learned is that the soil type makes a large difference in terms of speed of travel, slopes we can traverse. Meridiani is a totally different soil from Gusev. I could give you a number but the error bar would be larger than the number itself ;-)

Paolo

Thanks Paolo - let's just hope it will be a long while before we need to use those error bars!!!!

Just reading Scott's blog - today, funnily enough - http://marsandme.blogspot.com/2009/02/spirit-sol-48.html - talks about the work required to churn soil, he's comparing the first Spirit trench to Opportunities...

"Opportunity's hole, by contrast, was nearly 10cm deep, and they needed less than half as much energy to dig it. Which would tell me something about the differing soil composition between the two sites, if I were that kind of person.)"

That would bode well for Opportunity (if the floor of Eagle crater is like the rest of Meridiani) - BUT - the dunes may become a worse danger if we're dragging.

Doug

Looks like 1806 is another drive sol. Keep going wheels!

Alright, here's my prediction: Oppy will back on parking lot terrain (Aqua/Green) on May 1.

That's assuming a rate like what we've done from Sol1800 to 1806 and no major pit stops.

Graphic showing track due S from Sol 1806 position.
(The predicted ickiness of the dune field to the E of Porcupine varies between the different models - the most conservative is shown)
Click to view attachment
The black box at the end of the arrow shows the approximate location of first Aqua/Blue terrain on this vector.
(All the models agree that this would be "Aqua/Blue" terrain - somewhere between Sol1691 and Sol1693 in difficulty.)

-Mike

Left turn at some point? Maybe soon?
Click to view attachment

Given that 6-wheeling days may be numbered, what is the shortest route to permanent (as in all the way to Endeavour) parking lot terrain?

Turning E immediately and weaving through the flattest (blue/aqua) terrain to the green parking lot stuff.

Any reason to suggest why the RF wheel is the problem wheel for both MER's?

With all due respect, Mike, but I don't believe turning east now will do us much good. It will take us away from the 'yellow brick road' bedrock patches and on sand again, and it will mean driving perpendicular to the dunes ('against the waves').

Looking at the past weeks, it seems to me that trouble started building up when oppy was crossing a patch of sand, I'm still hoping that staying on bedrock as much as possible will help to reduce the load on the gears. This means following the 'yellow brick road' as set out by Rui/Ustrax, turning a bit more to the right, and keeping Porcupine on our left hand side, then making a turn east after passing Porcupine. There are still some patches you have to pass, but that will keep us on bedrock most of the time and 'nasty' dunes can be negotiated on a southerly heading which gives the least problems.

Click to view attachment

This is my own latest terrain calculation ('dune' filter) at 1 mtr/pixel for the area south till porcupine. Blue is suspect and Red should be avoided. The best option still seems to be to keep Porcupine on our left hand side (passing close to the edge, remaining on the bedrock). Other possible option would be to continue dead south from the 1806 position, keep Porcupine on the right and zigzag between the blue patches, however that would give a longer distance on sand.

Click to view attachment

The same for the area south of porcupine, once we have passed Porcupine terrain should get better and there will be a good option to turn east as soon as she is (temporarily) out of the dunes. This is conform the route of Rui/Ustrax.

Click to view attachment

Finally the same map again, but looking towards the east-south-east from the 1806 position. Turning east now will lead us away from the bedrock with several suspect 'dusttraps' and still a lot of dunes. Eventually it will lead us indeed to better terrain, but at the cost of having to drive perpendicular to the dunes for quite some time and on sand.

Presently I suspect Oppy is heading for this (as yet unnamed) buried crater a little over 100 mtr south of the present position, I guess the final route past Porcupine (passing east or west) will become clear once she is past this crater. We will wait and keep our fingers crossed regarding the wheels...

Regards,

Geert.

Dunno.

The pavement patches are starting to thin out in the "yellow brick road".

There are some potentially yukky dunes (magenta/red) in the dune field E of Porcupine. Even W of porcupine has some blue/magenta terrain.

I'm not so sure about the potential dust traps. Some of our models had predicted potential issues in the area near Sol1795, Sol1796, and Sol1797. It looks like an older infilled crater. We crusied through there with little problem.
The treads did sink in a little bit more than up in the parking lot near Endurance.

Turning E now, we would encounter terrain significantly LESS nasty than the recent cross-dune traverse of Sol1786 and Sol1791. Going ESE, within 100 m (2 driving sols), we would be in terrain less scary than Sol1691 - corresponding to the outer apron of Victoria Crater.
BUT, that cross-dune traverse across blue/aqua terrain would wiggle E for about 1 km before getting into true parking lot terrain.

From a science perspective, I'd imagine the drive E would be really, really boring. Aside from a drive by my favorite cute little 50 m crater, every day would pretty much look like the rest. ("Hey, look, more dunes.")

-Mike