That was my fault. (English isn't my mother language)
And now looking to the FHAZ images, the "trench" is really minimal. Perhaps we should rename the thread as "(Almost) stuck near Porcupine" or "Pit stop near Porcupine".

Now, going back to the first posts in the thread. At that point in time, with no available data except for the end of drive pictures, I was convinced the "incident" happened during the blind part of the drive. Later on I realized it was during the autonav part. I always thought the latter had "slip-check" (visodom?) enabled so such kind of excessive slippage would be immediately detected. Is that true?

Looks like exploratorium is back.

http://qt.exploratorium.edu/mars/opportuni...W6P2439L6M3.JPG

What a view!

Eduardo, English is not my primary language either, and I did not mean to criticize. It was really meant as a joke.

Slip checks are called typically every 10-20m, not at every step. Each slip check takes 5-12min, that's why we do not do that more often. We have determined that even if we have an embedding event whiel commanding 20m this should result in a situation that we should be able to resolve in a sol or two. This has been determined to be an acceptable risk. We do slip checks both during the blind (both forwards and backwards) and the autonav portion of the drive (forwards only).

This embedding evend does not surprise me. The lee side of the ripples are covered by a soft, cohesionless fine powder, while the opposite side is more compact and results in better traction. Autonav tried to avoid the ripple since it had a slope that was considered too high. The avoidance resulted into an arc that was sufficient to clear the ripple crest, but not the "fluffy stuff" (techincal term that drivers use to call this type of terrain).

So, IMHO, this is something that from time to time will happen, we have means to mitigat the effects and we know how to deal with the consequences.

Paolo

To be honest I think the whole event shows just how awesome the driving techniques are.

Yes - awesome is the right word!!

That is true. When I compare our current driving techniques with the first, say, 400 sols, there is quite a bit more safety checks and protection against all the things Mars has thrown at the rovers. This is without taking into consideration all the workarounds for all the aging actuators.

Paolo

This new view of the hills from today's pancams is a composite of three frames. We're starting to see a bit of detail on the hills.

Phil

Are these hills part of the big crater Oppy is headed towards?

Yes

Phil

Left + right, assuming the parallax at that distance is little enough:

You can almost hear it..

"Beeep...beeep...'Rover Reversing'...Beeep....beeep."

I am still wondering about the steering of those wheels, picture in last message is a good example of this. I'm aware that (with the exception of the stuck RF actuator) she is steering with all wheels, but how is decided which wheel makes which turn?

In above case the forward wheels seem to have made a nice shallow turn, with the center wheels following, however the left aft wheel seems to have made an almost 90 degree very sharp turn. Originally I thought she was slipping sideways but from the current angle it looks more like this was a deliberate turn, so why was the left aft wheel making a much sharper turn then all other wheels? Is this deliberately to compensate for the stuck actuator on the RF wheel, or did she get stuck first and then the left aft wheel slipped sideways as the right wheels kept pushing forward?

I can imagine that if the left forward got stuck with the right side continue pushing forward, the rover starts turning to left, this should be sensed by the IMU as a course deviation, so it starts compensating for this. Which wheels would it turn, or would it add more power to the left wheels? There is no sign that any attempt was made to turn the left forward wheel after it got stuck, and the left aft seems to have 'followed' the turn and not so much countered it.

Is there a system in place that notes that if you get a course deviation without a prior commanded wheel-steering this is a sign of trouble?

In a 'nominal' car, if you make a sharp left turn, the right wheels make significant more rpm then the left wheels, and the whole construction nicely compensates for this I guess MER does the same but in this case with the right side stuck that was counter active (same in a car, with one wheel stuck and the other side more or less up in the air, it's no use applying full throttle..). Is there is system in place to sense that one wheel is getting stuck, or is this only the slip-check every 20 meters?

What a mess you made, Oppy...!



(but we forgive you...! )

No - The FL, RL and RR wheels steer. The middle wheels do not.

We've already heard that slip-checks only occur every 20m or so, so it's not hard to imagine the vehicle getting 'stuck', but driving for another 15m, and crabbing sideways as it continued to push against the front wheels which were no longer making progress.

I think that one of the driving lessons are that the crest has the softest sand than the bottom between crest so the rover must avoid in rolling on that. The higher is the crest, the softer is the sand. I think.