Opportunity Sol 1045 (Spirit Sol 1063)

So now Opportunity is sitting in front of a rock known as Santa Catarina, ready to IDD the heck out of it. As has happened before -- and recently, too -- the target is heavily shadowed, so our terrain mesh is poor. There's a sort of "king's crown" of mesh spikes surrounding the rock and pointing toward the FHAZ.

This makes the whole day ... well ... interesting. Hand-estimating surface normals to the target is always dicey, and it's worse when you reflect that we can break the IDD if we're off by more than 15 degrees.

So that we won't have to do this tomorrow, we also make some time for swinging the IDD away and taking another FHAZ of the IDD workspace. If only the rover had fingers she could cross.

Meanwhile, Santa Catarina itself is an interesting rock. Well, not to me, maybe, but to the science team, and that's all it takes. According to Squyres, "Santa Catarina is the first rock we've seen at Meridiani where, a, we can say something about where on Mars it came from, and b, it's not made of blueberry stuff." Good news for them, then.

Another piece of good news is that I actually see improvement in Terry's IDD sequencing. Especially on a day like today, which is plenty stressful already.

I just hope I haven't spent all my luck on that. I'm going to need some left over for the FHAZ we're taking. Especially since Opportunity can't cross her fingers.

[Next post: sol 1068, January 4.]

Courtesy NASA/JPL-Caltech. Santa Catarina, in all its heavily shadowed glory. On the next rover, I want headlights, you hear me? Headlights!


Opportunity Sol 1032 (Spirit Sol 1053)

Yestersol's drive didn't go quite as planned. Opportunity drove through a small depression that increased her tilt to 7.44deg -- just above her tilt limit of 7deg -- so she stopped, which of course was the right thing to do. But she's actually fine, and ready to continue.[1]

The remainder of the drive is only about 8m, and that stops us in a perfectly safe position, about 2m from the rim. But there's a ritual now: any rim approach causes some people to panic, and after a lengthy discussion, we end up moving the final position back another 50cm. Which is how it always happens. It's just silly: we're not going to be wrong by 2m over a drive of 8m, and even if we were, we have reactive checks -- such as the tilt limit that correctly stopped us today -- that would catch our mistake before the vehicle was in danger. The 50cm pullback is just a compromise to end the discussion, nothing more.

What bugs me about this most, I think, is the fact that the extra 50cm of buffer is there not for any technical reason but to mollify the exaggerated concerns of people who aren't RPs. There, I said it. It feels good to have it said.

Tomorrow, no doubt, the ritual will be performed again.

[Next post: sol 1063 (Opportunity sol 1045), December 30.]

[1] One of the hazards of the job, I'm afraid. You always want to pick a tilt limit that's just a little higher than you think the rover will actually experience, so that an excessive tilt will mean she's strayed off course. But it doesn't always work like that: faulty data or other problems can lead you to choose a number that's lower than you should have. Fortunately, nearly all the time -- as in this case -- the problem with being overly conservative is just slight embarrassment. As long as I still have a rover to play with the next day, I can take it.


Opportunity Sol 1019 (Spirit Sol 1039)

Yestersol's Opportunity drive went well, so we're not going anywhere in particular today. Mostly, we're going to sit and take pictures.

But we are doing a bit of driving. In order to optimize our downlink for the weekend, we're just going to do a comm turn.

Steve Squyres raises an interesting issue with this. Back when we developed our strategies for driving with this IDD problem, one of the things we told NASA HQ was that we'd hover-stow (a.k.a. "thinker-stow") whenever possible, mostly to minimize the risk that the IDD would fail in the stowed configuration. However, we've kind of lost sight of that, preferring to do a full stow on every drive, because it's so much easier.

It is also, in more than one respect, safer. In order to drive in the thinker-stow position, we need to ensure that we don't drive any wheel over anything bigger than 3cm. Also, we can lose calibration when we do that, because the IDD bounces around more than when it's tucked safely below the rover's chest.

What ends up putting the kibosh on thinker-stowing for the day is not any of those considerations, but one more esoteric. The rover won't drive when it thinks the IDD is not stowed, and the thinker-stow position doesn't meet the rover's definition of a stow. So you have to poke a flag that tells it to consider the IDD stowed anyway. However, the command that does that hasn't been tested under the new flight software, and the potential downsides are significant. So we don't do it. I'll have to look into the matter soon and ensure that we can thinker-stow in the new FSW, but at least for today, things stay simple.

[Next post: sol 1053 (Opportunity sol 1032), December 19.]


Spirit Sol 1037

And off we go. Our big goal for the day is just to turn from a heading of 167 to a heading of 270, the first piece of our Esperanza drive. If everything goes exactly as planned, we'll end up straddling the shelf that was off to our right when we were at Winter Haven. In the best case, we'll also try to bump about 1m back along the shelf, just to see how that goes -- but if it doesn't get a chance to run due to time constraints, that'll be fine with us.

Happily, we're up to a whopping 357 W-hr. Well, that really is whopping, for Spirit these days. That number keeps climbing, and as it does, we'll be able to move more and more. And as drives like this one build our confidence, I hope we'll be able to drive more and more confidently.

[Next post: sol 1039 (Opportunity sol 1019), December 5.]


Spirit Sol 1034

On the road again ....

Well, not just yet. We'll be bumping in a couple of days, exercising those weary wheels. But we want to do some MIs first, and we don't want to do IDD work and driving in the same day -- driving our five-wheeled vehicle will be plenty tough without any added distractions. So we're doing the MI today, and then we'll bump Thursday and/or Friday.

The surface we're MIing has something that looks like lapilli -- they're these little glass beads formed in volcanic ash. Whether the objects we're seeing are actually lapilli or not is an open question; that's just what they look like.

To a geologist, anyway. To John Callas, they look like peas, and that starts a whole lot of comparisons between the features we're seeing and Twinkies, pancakes, and I can't even remember what else. I think we all must be hungry!

[Next post: sol 1037, December 3.]


Spirit Sol 1032

It's only sol 1026 on Mars, but we're planning sol 1032 for Spirit -- nearly a week ahead. These long APXS and MB integrations will do that.

Next week, things get more interesting on this rover: come Thursday or Friday, we'll try to drive away. We want a spot where we can see the MB "damage" from the winter campaign, and from which we can begin our attack on Esperanza.

Should be fun. But the only thing we're up to today is a MB-to-APXS tool change. And it was already done. Yesterday. By a shadow. Makes me feel damn near useless.

Can't wait for next week.

[Next post: sol 1034, November 30.]


Opportunity Sol 1009 (Spirit Sol 1029)

I'm not on shift today, so I decide to run an errand and go in a little late. While I'm running the errand, my cell phone rings. It's Al Herrera.

"Hey, are you on Lab?" he asks. That's never good news.


"Well, uh, Ashitey was scheduled to be RP-1 today, and he's not here. And he's not answering his cell phone."

Now, that's as unusual as all get-out. "I'll be right in," I tell him. Tara -- RP-2 today -- is there, and she can hold the fort until I arrive. But I'm awfully worried. Ashitey's as reliable as they come; if he's not there, he's probably dead in a ditch. And don't think I'm kidding.

The terrain mesh is just amazing. We're perched near the rim of Victoria Crater, looking out into it, looking across the crater at another promontory. We could damn near drive over there if we wanted to, and if we thought we could somehow descend safely into the thing from here.

Instead, we'll be hauling ass in the other direction, mostly away from the crater. The plan is to make lots of progress toward our next imaging position, but to end this drive well away from the crater, to give us room to perform some more testing of the new mobility flight software. If those tests go well, we'll be able to use even more advanced autonomous hazard avoidance.

The bad news for today is that we don't have PCAM in the drive direction. So I pull out a technique we don't seem to have used on this rover in a while: autonav. It makes me feel good just to be using it again. We back away from the crater rim using VO, so that we'll know for sure when we're 5m or so away. Then we drive blind up to the limit of our mesh -- only about 10m -- and switch on autonav for another 40m or so, a total of about 60m.

Not bad at all, really: I didn't even think I would get to drive the rovers today, much less that I'd do one of the longer drives we've seen for a while. I'm almost disappointed when Ashitey shows up during the day. It turns out the whole thing was a misunderstanding: they'd originally planned no RP activities today, and when they changed their minds, Ashitey didn't get the word. So that explains that, and at least he's not dead in a ditch! Which is good news -- and what's more, since I've already invested in the day, we decide it's best for me to just complete the shift. Bonus!

One of the reasons we can afford to do such a long drive is that our energy has been creeping back up. We're now getting 560 W-hr per sol. It's enough that they suggest to Steve Squyres that this could be the sol we resume doing overnight ODY comm passes.

Steve's excited -- but skeptical. Not because he doesn't believe it, but because, as he puts it, Colette has been playing "Lucy and Charlie Brown" with him -- teasing him with the promise of an overnight pass, only to yank it away at the last second.

"So I'll believe it when I see it," he mock-pouts.

That makes it all the funnier when Colette works it out so that we can do an ODY pass, and Steve gets really excited, and then something or other changes and she has to take it away again at the last second. As usual, Steve takes it with good grace and humor. I'm sure those screams of anguish are in jest, mostly.

I almost don't want it ever to work out. This way is more fun.

[Next post: sol 1032, November 28.]

Courtesy NASA/JPL-Caltech. A small part of the view from today. Glorious!


Spirit Sol 1024

Opportunity's drive stopped early; what was to be a 1.5m bump was a 1m bump. But they can see just fine from where they ended up, so they're staying there.

Spirit's drive didn't go as well as planned, either; we failed to stomp the outcrop. However, we ended up where we wanted to be, which on this vehicle is a huge success. We might try again.

But for now, we're OK with where we are, and we're up to some light IDD work: MI+APXS thisol, RAT nextersol.

Terry's the shadow today, and he needs more practice using the IDD, so I hand it over to him.

It's a long day. A long, long day.

[Next post: sol 1029 (Opportunity sol 1009), November 25 -- MSL's launch date, we hope!]

Courtesy NASA/JPL-Caltech. We might not have done everything we wanted, but we ended up in a cool place. Those platy-looking outcrops make tasty food for our robotic arm.


Opportunity Sol 1000 (Spirit Sol 1019)

I called dibs on the Opportunity "Now Planning Sol 1000" sheet -- and got it. So I have the set.

Dibs is a very powerful concept.

I shouldn't actually be working today, but Sharon, who's our team chief and therefore handles the schedules (among many other things), is a woman of her word. They changed when they were going to plan sol 1000 -- that was originally scheduled to be tomorrow, but for complicated reasons they decided to plan it a day in advance, so it's getting done today. Sharon had promised me I'd get to work on sol 1000 on both rovers, so as soon as she heard about the change, she came to me to alert me to it, and see who I wanted to bump so I'd be on shift.

Jeng and Antonio were RP-1 and RP-2, and Tara was shadowing. So who did I want to bump off the list? Well, what kind of a person would I be -- what kind of a team lead would I be -- if I took away the fun of planning sol 1000 from someone else? So I had Sharon add me to the list as a second shadow.

And here I am, shadowing, as we drive Opportunity to the rim of Victoria once again. It's a short drive, and one we can do without a lot of uncertainty, using VO all the way. But certain people (Cindy Oda) have been freaking out whenever we drive up to the rim, and today is no exception.

So today I try a new approach. I start trying to compare the drive to things people can see around them. We're driving less than the width of the room we're in, a little more than the length of our conference table. I measure the size of the carpet squares; they're a little more than a half meter on a side. Walt Hoffman has a great idea: we've got a couple of 3m network cables lying around, which gives you a handy way to measure 3m (our distance from the rim), 1.5m, and 1m, at least.

I think this helps. It gives people something more concrete -- being able to say "We'd have to be wrong by the length of this 3m network cable, on a drive the length of this conference table" is pretty effective.

So everything proceeds apace. Until John Callas shows up late in the day and starts asking the same questions all over. And even that is going fine until Justin Maki -- a level-headed, bright, and valuable guy -- starts winding Callas up. "Well, what if we had two meters of range error in the NAVCAMs you're using for the drive?" Justin asks. Two meters of error? This is just absurd, and very much out of character for Justin. A realistic number for range error over that distance is maybe 8cm.

But John Callas doesn't know any of this, and it's his ass if something bad happens. So that's a painful conversation.

John Wright is listening to all this, and doesn't say much. Except, when it's all over (and we've cut the drive short another 50cm for no good goddamn reason), he tries to make me feel better about it. In his way. "Spirit's probably not going to survive another Martian winter," he points out. And he's right, we barely made it this time. "So keep Opportunity safe," he continues.

Yeah. I see his point. I mean, we were keeping her safe before, but ... yeah. I see his point.

Oh, well, we're still going to enter sol 1000 with a hell of a view. As is his wont, Jim Erickson pops by to see how things are going. "We're planning sol 1000 today!" I tell him brightly.

He looks down his nose at me, mock-snide. "Talk to me when you get to sol 10,000," he says.

Don't think I won't, Jim!

[Next post: sol 1024, November 19.]


Spirit Sol 1007

It's my first day back on shift after conjunction. Both rovers weathered conjunction just fine -- maybe they don't need us as much as we think they do!

During conjunction, with the sun firmly planted between our two planets, Spirit also survived crossing the sol-1000 barrier. And I managed to snag the "Now Planning Sol 1000" sheet, and got everybody to sign it. Sweet, sweet victory. Since I'd requested to be scheduled on sol 1000, Sharon scheduled me for that day, even though there was no planning that day. I just walked around all day with a big dopey grin on my face. It'll be even sweeter when Opportunity's sol 1000 comes around in a couple of weeks.

This will be the first time we've MId Spirit's solar panels since about sol 505, apparently. Thisol is -- or, rather, these sols are, since the work executes across 1006 and 1007 -- pretty MI-heavy. Ken Herkenhoff, the MI's designer and lead, is going to be in heaven.

Meanwhile, here on Earth, not everything is going that well. Our building has a partial power outage, leaving us with nothing but emergency lighting and the glow from our monitors. I love it, actually. It's all Star Trekky. This is how things are supposed to look around here!

But there are also problems with Maestro, our planning tool, that threaten to delay the day significantly. Fortunately, our TUL, Colette, manages to get things under control, and we finish the day on time -- not with a lot of margin, but some. She's a heroine!

[Next post: sol 1019 (Opportunity sol 1000), November 14.]


Opportunity Sol 961 (Spirit Sol 982)

Today we're continuing the pre-conjunction IDD activity to execute during conjunction. In one of those wonderful happy-fun-time events, there are two RAT targets -- and neither of them really works, thanks in part to wonky data in the terrain mesh caused by shadowing. We end up taking another image of the currently shadowed area, with modified image parameters that we hope will improve range data, and in the meantime we do what we can with what we have.

That turns out to be a huge pain. When at last we get something that seems to be working, we run into a joint limit on the APXS and have to start looking around for alternatives. But the alternatives that are RATtable have even worse problems, so we go back to the first target. Eventually we tweak the normal so that we can place all instruments there, but the tweaked version can't handle more than about a 10N preload from the RAT. That's not nearly enough to brush, to say nothing of grinding -- we need 40N or so. So we keep working on it ....

Eventually, we end up with a two-pronged strategy. Prong one is to choose a target in the shadowed area, despite the poor range data there. Prong two is to cross our fingers.

For the first time in a long time, there's a press conference playing on the TV behind us as we work. It's them science guys, talking about Victoria Crater. One of the images they show illustrates the scale of the crater -- they show the Rose Bowl fitting into it, with room to spare.

Wow. I see why the science team wants to stay here a while. So much to see!

[This is where we had to take a few weeks' break for solar conjunction: a giant flaming ball of gas (no, not Rush Limbaugh, the sun) interposed itself betwixt us and our beloved rovers. That made radio communication impossible for a while and thus interrupted normal planning. Next post: sol 1007, November 2.]

Courtesy NASA/JPL-Caltech. Victoria Crater swallows the Rose Bowl.


Opportunity Sol 961 (Spirit Sol 981)

The drive went great. Today's all about the IDD: MI, RAT-brush, MI, APXS on the first sol; then RAT-grind, MI, and APXS again on the second sol.

Piece of cake.

Courtesy NASA/JPL-Caltech. Perfect driving put this patch of rock dead in front of us, right where we wanted it. RPs, one million; Mars, zero.


Opportunity Sol 959 (Spirit Sol 980)

The drive went beautifully, or so we think. We got a nice PCAM of the slab of rock we're targeting, just 6m away. The bad news is that we didn't get the HAZCAMs for some reason. One of the other RPs floats the idea of driving even without them, prompting Steve to tease, "Are the PIs more conservative than the RPs now?" But, I say, "Nope: we don't drive without the HAZCAMs."

Steve agrees. "We drove aggressively so that we would have a margin day. Let's use it."

That becomes the default plan. But it gets under my skin. We worked hard on yestersol's drive, and it worked, damn it. I hate being stymied by something like this. So I go down to MDOT and ask them what happened to our data. Sometimes it didn't make it to the spacecraft for one reason or another, but sometimes it's just a ground processing problem of some description -- a server needs to be restarted, or something. It's worth asking, especially since I won't have anything else to do if the HAZCAMs are truly missing.

The answer from MDOT is that the data made it from Opportunity to Odyssey (our relay orbiter), but they're not sure yet what happened to it from there. It's either still on Odyssey, in which case we won't get it until tomorrow, or it's in the Odyssey TDS (the telemetry data server), in which case they might be able to figure out what's wrong and get it unstuck.

This is actually pretty good news, as far as it goes, so I take that back upstairs with me. "Change of plans," I say. "Let's plan as if there's going to be a drive. If there isn't, we'll pull it and just do remote sensing."

Then I turn to the RPs. "While we're waiting on the word from MDOT, let's see what we can do without those HAZCAMs. The burden of proof has to be on those who want to drive: we'll have to prove it's OK to drive without them. But let's bring every piece of data we have to bear on that. We've done it before, but we know we'll have to have a wider discussion -- especially since we're so close to the rim -- and we should have a case to make when that happens." They swing into action.

Man, I'm on fire.

And it turns out it's a good thing. For some reason I don't quite get straight, we lost our uplink pass tomorrow, so today is our last day to bump before conjunction. (Well, we have one more day if we absolutely need it, but we really don't want to spend that one if we don't absolutely have to, since we're not sure exactly when we lose contact with the rovers, and we want to get some IDD work in first so that we can start a long integration over conjunction.)

And then we get the word from MDOT. They found our HAZCAMs! They've been here on Earth the whole time. They've restarted the server and we should get the data in about 20 minutes.

The good news is that when the images pop out of the server, they show nothing alarming. Just as we expected -- and we can drive! The bad news is that more data pops out at the same time, and it shows an unexpectedly high current draw on the RF wheel. No telling why, and we decide there's not much we can do about it. We're driving.

The actual sequencing isn't my concern, for the most part. I'm not on shift today, just hanging out to keep an eye on things for this near-rim drive. (Ah, the responsibilities of being a lead RP!) But it's a straightforward one, appropriately simple: we drive up to a rock target ("Fogo"), bump onto it to scuff it, and then back down a bit so that we can see the scuff and IDD the rock.

(This mini-story doesn't fit anywhere, but it's a classic. While we're chatting about the sequence, my cell phone rings, playing a Smiths song, and Matt Heverly grins at me. "You're the only adult I know who has that many ring tones," he quips. "You should be a junior high kid.")

Speaking of data coming down during the process ... while we're looking at the drive during the APAM, Squyres pipes up with an interesting question. "Hey, we just got the HiRISE image of Opportunity at Victoria Crater," he announces. This is the high-resolution MRO image we've been eagerly anticipating. "Would it improve the tactical planning process if I --"

"Yes!" I declare without even thinking about it.

"Whom should I send it to?" he asks.


I really am on fire.

A few moments later, we're looking at Opportunity, as seen from orbit. She's sitting there, perched on Cape Verde -- just a few pixels across, and most of that mast shadow, but it's undeniably our baby. This is astonishing. It makes her real all over again. I can't even find the words.

We're going to take good care of her.

Our sequence is appropriately cautious, sporting all the bells and whistles that it should for something that takes us so close to the rim. And it helps that we're going uphill to the rim from here, so if we slip at all, it will be in the safe direction, away from the edge. But everybody's nervous about it anyway. As Matt Heverly leaves, he sums up the feeling: "Well," he says, "see you at the Congressional Anomaly Review Meeting!"

Courtesy NASA/JPL-Caltech/University of Arizona/Cornell/Ohio State University. Steve sent us a black-and-white preview of this glorious HiRISE (MRO) image showing, unbelievably, Opportunity from orbit. Click through for the full, spectacular image, available in both annotated and unannotated versions.


Opportunity Sol 958 (Spirit Sol 979)

"Have you see Mark Maimone yet?" Matt Heverly asks me as I walk into the sequencing room. "He's kind of freaking out."

And he's overreacting. The drive went just about exactly as planned, leaving us poised near the rim with a beautiful view out onto Cape Verde. It's the "near the rim" part that Mark was apparently freaking out about. By the time I find him, he's calmed down, admitting that he'd misread part of the uplink report.

So if he's happy, I'm happy. Matt and I go off and plan thisol's drive, one that will take us 20m out onto Cape Verde. If this one goes well, it'll leave us set up for the final bump to our pre-conjunction imaging position. And nobody will be freaking out.

Courtesy NASA/JPL-Caltech. Peeking over the rim of Cape Verde into the interior of Victoria Crater.


Opportunity Sol 957 (Spirit Sol 978)

So all we have to do is approach Cape Verde. It would help if we knew where it was: we have dueling localization estimates, and in the end we take our best guess.

Thanks to the recent flight software upload, we're nervous about using our usual waypoint commands anywhere near the crater. So we sequence a 40m shot at Cape Verde, using lower-level driving and reinforcing turns every 5m. The last 5m uses guarded motion -- it will autonomously stop if it detects a hazard, such as a cliff edge -- but otherwise there's nothing fancy.

It's gonna be a hell of a view. Victoria Crater is an awesome place, well worth the slog it took to get here.

Since we're shooting toward the crater rim, we decide to put in a few keepout zones to make doubly (actually, triply) sure we don't go in. Since circles are the easiest kind of keepout zone to specify, we put a big circle directly in our path, and two smaller ones off to the sides. I take a look from the overhead view in RSVP ... then adjust the two smaller circles slightly, and it's perfect: the Mickey Zone of Death.

Courtesy NASA/JPL-Caltech. NCAM mosaic from sol B-957. Spec-freaking-tacular.

The Mickey Zone of Death!


Opportunity Sol 943 (Spirit Sol 964)

The bad news is, we're starting at 7 AM -- which means that I and the MM and TUL and SOWG chair and a few others have to start even earlier, about 6:30. The good news is, this is the last time we do that, ever. Our much-loathed 7 AM starts are now a thing of the past; no more starting before 8. Well, those of us who have to be a bit earlier will start at 7:30. But it beats 6:30, I'll tell you that.

Since I can hardly focus anyway, it's a good thing today's drive is a simple one. We plan 57m, dead at Victoria Crater, stopping midway to image a smaller crater called Kitty Clyde's Sister. This drive cuts our distance to Victoria roughly in half: we should be about 51m away when we're done.

From 108m away, the 800m-wide Victoria crater is the broadside of a barn. Even starting at 6:30 AM, we couldn't miss it with our eyes closed.

[Next post: sol 978 (Opportunity sol 957), October 3.]


Opportunity Sol 938 (Spirit Sol 958)

I'm not on shift today, but this is something I couldn't leave out. As part of my investigation of the Jammerbugt anomaly, I suggested that we reinstate the practice of having the science team periodically brief the engineers on recent science findings. Today's the first installment of that, being kicked off by Steve Squyres himself.

There's way too much to summarize here -- and I didn't bother taking many notes anyway, since they were filming it -- but a couple of things stood out for me:

"Mars is a terrible place -- the average daily temperature is -60C. All the water vapor in the atmosphere would form a layer a hundredth of a millimeter thick."


"Meridiani and Gusev were the two best-studied places on Mars before landing. We had Odyssey, MGS, Viking data. And we had 'em both completely wrong! We expected volcanic rock at Meridiani and sedimentary layers at Gusev."
(What we found, of course, was basically the reverse. Shows the value of surface data to complement orbital data!)

[Next post: sol 964 (Opportunity sol 943), September 19.]


Opportunity Sol 931 (Spirit Sol 954)

We're still not out of the dead of winter, and thus still not out of the woods. Daily energy levels aren't rising -- but they seem to have shallowed out, at least. At the same time, tau (atmospheric opacity) has spiked, and a more opaque atmosphere is bad news for solar-powered rovers.

However, tau has been on the low side lately, and is now only about average for this time of year, so it's nothing to worry about. As Oded points out, we mustn't panic about one measurement; we should wait to see if it's a sustained spike and one that causes power to drop.

And what are we going to do about it if it is rising, anyway? Go fly around with a vacuum cleaner and hoover the Martian atmosphere? No, we'll have to imitate a vacuum in another way, and just suck it up.

[Next post: sol 958 (Opportunity sol 938), September 13.]


Opportunity Sol 925 (Spirit Sol 946)

Well, the tests on the IDD have gone well, and yesterday they even did real work with it. Maybe it was just a transient failure, after all.

However, we're still nervous about it, mostly because we're in for pain if it fails while we're trying to stow. During the stow and unstow process, the IDD maneuvers itself through a number of tight spaces, and getting it back out can be a real bitch. And stowing contains several small joint-1 moves, so if we didn't just have a transient failure -- if we have a real, ongoing problem -- then it's possible we'll fault out when that happens.

Fortunately, we can do something a little more active than just cross our fingers. We can walk the IDD partway through the stow process manually, at least as much as is relatively easy to sequence -- and to undo, if we get stuck. That will either build our confidence that the real thing will succeed, or tell us we shouldn't bother trying the real thing.

Happily, Eric Baumgartner left behind excellent documentation, including the exact IDD joint angles we need in order to do this test. So we look those up, put the sequence together -- and get shot down. Well, not permanently, but the SOWG chair's call is that since our doing this would put thisol's MB observation at risk, we should punt. And I can't really disagree, since we have a chance to do it again tomorrow. So we put our nearly complete sequence on the shelf, point the next day's RPs to it, and move on with the rest of the day's IDD sequencing. At least it won't be work wasted.

[Next post: sol 954 (Opportunity sol 931), September 9.]


Opportunity Sol 921 (Spirit Sol 942)

Grim news from the Land of Opportunity today. The drive's off, because yestersol's IDD work on the trench failed. On the very first IDD move of the sol, we got a joint-1 stall at 80 ohms, which we haven't seen before. And the most likely interpretation of that is that the IDD has continued to degrade, and we're about to see a whole lot more failures. We might have lost the shoulder joint altogether. Jake Matijevic looks unhappy, which is never a good sign. "I'm concerned whether we'll ever be able to reliably stow again," he mutters to me sotto voce.

The only ray of hope is that the 80 ohms value isn't necessarily a magic number. It's a number we picked because it was supposed to work 90% of the time, or something like that. And if this is the first time it's failed in lots and lots of tries, maybe we just had some kind of bad luck, a purely transient failure due to ambient temperature or something. It has been particularly cold lately.

And Kirk Fleming makes a good point. "The first move of the day can be kind of sludgy," he notes as we're sitting in the fishbowl reviewing the data. "The stall doesn't seem to be resistance-related, or we wouldn't have hit the current limit. So either something else is wrong with the same actuator, or we just had a transient glitch."

Well, that is good news, and real reason for hope at last. Anyhow, we'll know more tomorrow than we do today. We're planning a test of the IDD in which we do a series of small shoulder-az motions, at increasing resistance values, until we get a success. That, of course, is hoping we do get a success, which is by no means assured at this point.

After a lot of discussion, we end up with a fairly simple scheme. We're going to repeat a test at three different times of day, so that we can test at different temperatures. Each test moves the IDD to the left and right with a rotor resistance of 80 ohms, then 90, then 100, the max we can command. If we succeed in moving it both directions at a certain resistance setting, we don't try the next-higher setting, since that's potentially dangerous.

And we'll take lots of pictures as we do it. Lots of pretty pictures. Which, we hope, will -- like the other data -- show nothing anomalous. That would leave us, perhaps, with a mystery. But given the alternatives at this point, it's a mystery I think I could live with.

[Next post: sol 946 (Opportunity sol 925), August 31.]


Opportunity Sol 919 (Spirit Sol 938)

I take a personal satisfaction in today's activity: trenching with a stuck wheel.

Sure, I'd rather be high-tailing it toward Victoria. In my absence, the team got us to Beagle and beyond. Now nothing stands in our way -- well, nothing except the science team, who feel there's some science work we really need to do first.


The good part is that we're already prepared to do what they want to do. A few weeks before we reached the annulus, I got to thinking about that big old unknown sand sheet, and what we'd want to do once we got there. Trenching was one thing I knew they'd want to do, and I knew we hadn't done it since Opportunity's RF wheel lost the ability to steer, which makes our usual trenching sequence unworkable. So I got Paolo, Jeff Biesiadecki, and (bless his heart) Rob Sullivan to go down to the testbed and work out how to trench all over again.

Squyres said -- very, very politely -- that this was a waste of time. But I had a feeling it would prove otherwise, and it looks like I was right and he was wrong.

Well, it had to happen eventually.

So that makes the sol easy, at least: the trenching routine's already fully worked out, and we just have to send it to the spacecraft.

But you knew it wasn't going to be that simple, right? We're actually finished with the CAM when I realize what's wrong. The new trenching sequence draws a long, shallow scrape with the RF wheel, then turns slightly and makes a second scrape overlapping the first, thus broadening the trench. Now, the turn between the two scrapes is performed in an unusual way. Rather than turning through an explicitly commanded angle, which causes the rover to turn until the IMU -- roughly, her internal compass -- says it's time to stop, the mechanism used in the sequence simply runs the wheels a certain number of revolutions.

When I change this to an explicit turn for simulation purposes, the error is clear: we're turning way, way too far.

I call Paolo and confirm the problem. They worked this sequence out in the testbed, and the number of wheel revolutions to use in the turn was calculated empirically. In the testbed -- in the particular soil material in use there, in Earth gravity, and so on -- this does the right thing. On Mars, it might do the right thing, or it might turn the rover way too far.


The change to the sequence is simple enough; we just have to replace the turn with an explicit one. I'm glad I caught this, but, boy, do I feel dumb for not catching it sooner. My first day back from England, and we're redelivering post-CAM.

Just like the old days, really.

[Next post: sol 942 (Opportunity sol 921), August 27.]


Spirit Sol 912

"Aren't you gone yet?"

That's Alicia Vaughan (nee Fallacaro), asking me the same question everyone else is asking me. It's enough to make me feel unloved, or at any rate it would be if they weren't asking when I was leaving for England.


Contrary to my hopes, we obviously won't be at Victoria before I hop the pond and lose touch with all this for a few weeks. Heck, we're not even to Beagle yet. Now I have to hope they don't get anywhere interesting until I get back.

It almost makes me want to sabotage thisol's sequences, just to give myself some extra margin ....

Heck, yestersol's sequence was nearly sabotaged for me. Yestersol's RAT calibration was undermodeled, so the IDD sequence cut off at the very end. It had taken the final image, but hadn't compressed the IDD data products yet. So we didn't get those. As a result, we don't have detailed current data, but we do have a visual OK and initial conditions, and that's good enough for us to proceed. So we'll clear the errors that resulted from the premature sequence cutoff and then continue down Rob's Yellow Brick Road.

We're off to see the wizard -- and I'm off to see some Shakespeare.

[Next post: sol 938 (Opportunity sol 919), August 23.]


Opportunity Sol 890 (Spirit Sol 910)

The plan for the day is to drive on toward Beagle. First, we're going to repeat the unsatisfactory scuff done on sol 885, this time recording motor/IMU data at 8Hz to better characterize the rover's interaction with the terrain. We'll take juicy color PANCAMs of it -- or just get decent-quality HAZCAMs if we can't afford the PANCAMs -- and then we press on.

We're all fired up for driving, too. Tim Parker's "Victoria Goal Map" is showing us so damn close! We're only 85m from Beagle, and 561m from Victoria. I expect we should be able to do an easy 50m or so thisol, which would put us one drive away from Beagle and cut our distance to Victoria by about 10%. I am totally jazzed about this. We've got it all worked out, and we're raring to go.

That's why I'm so disappointed when the drive is cancelled. During the SOWG meeting, the science team decides the IDD data they've gotten at this location isn't quite high enough quality, and they need one more sol before driving on.

Bummer. But at least I get a good laugh out of Rob Sullivan's description of the now-irrelevant work we did on the scuff: "a fabulous exercise in futility."

[Next post: sol 912, July 27.]


Spirit Sol 908

The plan for the day starts with two 7-deep MI stacks of ripples Spirit hasn't paid too much attention to yet. The stacks are 7 images deep rather than our usual 5 because they don't want us to disturb the soil by touching it with the MB first. This is the start of a longer campaign to get MI coverage of a strip all the way between the two ripples -- a strip I soon begin to call the "Yellow Brick Road," because Rob Sullivan indicates its position with a yellow line in a planning image. The two 7-stacks are all we can afford today, but there's more in our future.

So I have what I think is a clever idea: since we know where the best focus position is likely to be, why not take a gamble? If we did five 3-stacks along the whole Yellow Brick Road, centered where the best-focus position is likely to be, then one of two things would happen. Either we'd get the best-focus images all the way along, in which case we win big, or we get enough information to work out where the best-focus position definitely is, and we come back and do a single strip of images at that best-focus position later. Either way, it's many fewer images than the planned method, and we might even get done in a single day.

Turns out, though, that this is the wrong time for the creative approach. In this case, Rob actually wants the whole 7-stack, including the out-of-focus images, because the terrain has some relief that he wants to be able to study in detail. Best-focus in one image is out-of-focus in another in a case like this. So that brilliant idea is off the table -- this time. But maybe next time.

So we get started down the Yellow Brick Road. (In keeping with the theme, I naturally push for calling our intermediate targets "Dorothy" and "Tin Man." Whee, the fun we have!) As we go along, we keep in touch with Rob, who requested this observation and is now calling in from the road on the way to a family reunion. We get to reminiscing about the nominal mission, when Rob -- one of the few who was even more hard-core than I -- practically tried to stay up 24 hours a day for the whole thing. "I used to roll from rover to rover and just pass out on the floor in the MI room," he laughs. "Hey, we should gather pictures of everyone sleeping!"

Sometimes I kind of feel sorry for the new kids -- such as Antonio, who's shadowing today -- who missed out on the overwhelming fun and excitement of the nominal mission. But at the end of the day, I overhear him on his cell phone, excitedly telling his wife that today he wrote his first sequence for the rover. The bloom isn't off the rose yet.

[Next post: sol 910 (Opportunity sol 890), July 25.]


Opportunity Sol 875 (Spirit Sol 896)

So close, oh, so close to Beagle. Less than a hundred meters now. But Mars isn't making it easy for us. Around Beagle, there seems to be an earthen (marsen?) rampart, a huge wall of impenetrable ripples. As far as we can see, north to south, the darn thing is just too tall to get over.

And if we just got over this one ripple, we'd have pretty smooth sailing all the way. On the other side lies the Beagle Highway, a sizable stretch of outcrop we could follow all the way in.

Yestersol, they tried the obvious approach: just power over this damn thing. It was risky, but they were careful -- still smarting from our recent experience in Jammerbugt, very careful -- and therefore sequenced quite conservatively.

And it was a good thing they did. We ended up getting partially buried in the ripple, almost as soon as we got fully onto it. But, because of the previous RPs' conservatism, we're not so badly buried that we can't simply back downslope.

Which is our plan for the day: just back away slowly. It's a short drive, but not a short day: as someone points out, discussion = 1/distance.

Before we go too far down, though, we're going to take advantage of the fact that we're about as high up as we're going to be for a bit: we'll snap some pictures of possible alternate routes to Beagle from this relatively high vantage point. We hope they show us something good, because we haven't got a whole lot of promising routes at this point.

By which I mean, obviously: we haven't got any. We're hoping the images will show us at least one.

Of course, there's always this option: just bypass Beagle altogether, get onto the sand sheet where we can move freely, then loop around and come at Beagle from the south instead. Rampart, schmampart: treat the giant ripple as a Maginot Line.

After we're pretty much done for the day, our old pal Jim Erickson stops by for a provocative discussion. "What tools would you want to have to drive the next generation of rovers?" he asks. I think he's just giving me an opportunity to plump for using RSVP on MSL, but he's after more than that.

So far, he points out, we've basically kept single drives within a known range of terrain types. When terrain types change significantly -- e.g., when Opportunity went from flat plains to the ripple terrain -- we've been able to adapt slowly, even to the point of uplinking new flight software to deal with the terrain changes. What he wants to know now is, how can we go beyond that? How would we have to go about driving a rover if a single drive might carry it across multiple terrain types, some of which we hadn't seen yet?

Hmmm ... now that's thinking big. I think we'd want some way to recognize and categorize terrain types, then have the rover switch to different strategies based on what type of terrain it thought it was in. We'd have a collection of strategies for known terrain types; for truly novel terrain types, it might have to go around or wait for help.

That implies a lot of changes to RSVP. Time to get to work.

[Next post: sol 908, July 23.]


Spirit Sol 893

It's a fairly slow day. We're just carrying out some RAT diagnostics on Spirit, picking up the IDD and posing it for a few images of the RAT before putting the MB back where we found it.


The IDD work we're doing will take place on sol 893 -- the second sol of a two-sol plan. On sol 892, Spirit will simply nap pretty much all day. And it'll keep getting worse for another three weeks or so, until the winter solstice. Meanwhile, we don't seem to be putting as much data into flash as we can downlink. We can't afford to -- the science observations that would fill up flash are just too costly in terms of energy.

Spirit is a sleepy basset hound, snoozing on the front porch while flies buzz around her head. I plan to work her as hard as possible when we can. But for now, we'll let her get her rest. I hope she enjoys it.

[Next post: sol 896 (Opportunity sol 875), July 11.]


Opportunity Sol 863 (Spirit Sol 882)

I'm in a little early and run into Oded Aharonson, whom I haven't seen in -- well, it's been a while. He's been off in Greenland or Iceland or Norway or something, working on Cryobot, some kind of prototype ice driller they might send to Mars someday.

Which is why I haven't seen him around, of course. Being away from MER seems to have given him a certain perspective on what we're doing -- "shepherding the rovers to their death" is how he puts it. He doesn't mean it in a bad way, and I know exactly what he means, but I have to confess I'm still in denial about that. They're gonna live forever, ya know? Just a few aches and pains, is all. They're fine.

Speaking of those aches and pains ... "How are we at five-wheeled driving?" he asks. "Can we get back up onto Home Plate?"

"I doubt it."[X]

"Circumnavigate Home Plate?"

"That's more likely. Since you bring it up, what's the obsession with Home Plate, anyhow?"

Oded's always animated, but a question like that always stirs the pot. "Home Plate," he explains excitedly, "is the first time we've seen a geological feature we have a chance of explaining, a chance of forming testable hypotheses about. It's not water, but it's still interesting." Ah-hah ... looks like even Oded thinks there might be a little life left in this mission.

Though you might not know it so much by today's plan. We didn't receive sol 860's PCAMs, so we're stuck planning a drive in the NCAMs, and that limits us to about 20m. We don't have any significant IDD work, either, though we do have a bit of IDD checkout at -- oddly enough -- the end of the drive. This being Opportunity, we have our normal post-drive unstow. And after that, we stretch the IDD out a little further to check out a problem with the MB's reference channel.

This is a data stream that feeds calibration data from the MB's internal calibration target into the flight software, so that we have calibrated measurements from that instrument. For some reason, we seem to have lost that data entirely -- possibly some kind of cabling problem. All hope is not lost for that instrument, since we still can place the MB on the rover's external calibration target (just above where the IDD is stowed). But it's one more of those little aches and pains.

The rovers are fine, you know. Just fine. And they're going to stay that way, forever.

[Next post: sol 893, July 8.]


[1] Wrong, of course. Didn't I know better than to bet against Spirit? The next driving season came around, and up onto Home Plate she went.


Opportunity Sol 857 (Spirit Sol 878)

Sure enough, we got about 40m out of the previous drive -- not yestersol, but the sol before, since we're in restricted sols. We're now about 776m to Victoria, and only 304m to Beagle. Thisol, we're gonna step on the gas a little, and shoot for about 50m.

We've got a guest driver helping us. His name is Kelly Wills. He's spending the summer at JPL, working on an MSL task -- trying to reduce the power requirements of one of their instruments. He's from Ohio, he just graduated from high school, and he's blind.

I try to remember what I was doing the summer after I graduated from high school. I'm pretty sure it wasn't anything like that, though. And I'm not even blind.

Man, do I suck.

This somehow gets Paolo and me talking about our other, non-MER tasks. I remind him that he's not allowed to leave MER, and he assures me there's no danger of that. "There will be four people here to switch off the last workstation," he says. "Mark Maimone, Steve Squyres, you, and me."

I can't speak for the other three, but he's right about me, at least.

Speaking of Steve Squyres ... he's our friendly SOWG chair today, and something happens today that makes me suspect I'm not a complete waste of water and trace chemicals. During the B-833 anomaly investigation, it came to my attention that several of the RPs, particularly newer ones such as Paolo, had trouble standing up to pressure from the science team, particularly Steve. In my new role as the RP team lead, I urged them to stand up to this pressure, pointing out, among other things, that Steve pushes and expects you to push back when he's pushed too far -- and you shouldn't disappoint him.

So, at the APAM, Steve, smelling a long drive, says something about "putting in as many slip checks as duration allows." And Paolo says simply, "I think I will use a different philosophy: we'll use as many slip checks as we need to be safe, and we'll let the distance be what it is."

I coulda kissed him!

Oh, and speaking of water and trace chemicals .... (I really am the King of the Segues!) Steve also relates a recent science find. Concentrated calcium salts, plus zinc, plus hematite, all in one location -- the implication being, maybe, hydrothermal hot springs.


The other news is not so good. Opportunity's MB has lost its reference channel, which carries internal calibration data. That doesn't make the instrument useless, as we can still calibrate it against the external calibration target mounted above the IDD's stowed position. But it raises a disturbing possibility, one that Matt Heverly points out: combined with the recent transient failures we've seen when commanding the MI, it might be that we're seeing a degradation of the flex cabling that carries power and data along the IDD. If that failed, we might lose the IDD a whole lot sooner.

Fortunately, Matt came to JPL from ASI -- the company that built the IDD. And he knows the guy who was in charge of the flex cabling. "I'll shoot him an email," Matt says.

I love that. If you heard a knock coming from under the hood of your car, wouldn't you love to be able to just email the guy who built the engine?

[Next post: sol 882 (Opportunity sol 863), June 27.]


Opportunity Sol 855 (Spirit Sol 876)

We're continuing to make good progress toward Victoria. Close enough to taste it, or anyway smell it. Or maybe just hear it.

Anyway, we did 40m yestersol, and we'll do 40m again thisol. This one's a little weird; we're driving up onto a sort of curb. We see these once in a while -- abrupt little steps in the terrain, usually, as here, associated with rock.

Why do they happen? I don't know, I just drive over 'em.

But not without some paranoia. The curb is at the limit of our ability to precisely measure, so after some discussion, we end up loosening the suspension limits considerably for driving over the thing. It's not implausible that we'd pop a wheelie going over it -- this happens when one of the bogie wheels (usually the leading wheel) loses purchase but its partner keeps going, as can happen when you're driving over a feature like this. The suspension auto-corrects within a meter or two, and no harm done. We wouldn't want to cut the drive in half if that happened, so we effectively tell Opportunity to just ignore it if it does.

And once again, we're plodding toward Victoria.

[Next post: sol 878 (Opportunity sol 857), June 23.]


Opportunity Sol 852 (Spirit Sol 873)

Opportunity's entering restricted sols -- "a good thing, for once," as somebody points out, "since we don't have enough power to drive every sol any more anyhow."

We're losing energy, and worse, we're losing Eric Baumgartner. (I am the King of the Segues.) He's heading off to become the Dean of Engineering at Ohio Northern University. I hand him a poster we all signed for him, and shake his hand.

Eric is the kind of genius I can only wish I were. Smarter than hell, a fantastic engineer, but also great with people -- terrific social skills, a top-notch manager. And somehow he finds time to coach his kid's soccer team. It's a real loss to the Lab.

Depressed, I head downstairs to plan thisol's drive, which is a challenge. The trough before us ends in sort of a cul-de-sac about 20m away. Lying across the trough is a sort of "pitcher's mound" -- likely soft material we don't want to drive over -- and a few meters behind that is another one. Paolo wants to stop short of the first pitcher's mound, but I think we can take it, and I manage to convince him of it. Instead, we're going to drive up onto the ripple beside it, nestle in the hollow between mounds, and then scoot off into the next trough to the east. If we make it that far, we'll turn south again and back across a big patch of outcrop.

I'm either insane, or a genius. I mean, not an Eric Baumgartner genius. But a genius, anyway.

Or insane.

[Next post: sol 876 (Opportunity sol 855), June 20.]


Spirit Sol 872

The bad news is, we're under 300 W-hr for the first time -- 297 is the actual number. 250, of course, is the lower limit of survivability (at least without pulling Opportunity-style Deep Sleep tricks).

The worse news is, they don't need rover drivers today. So I sit down with Tara Estlin, who's had bad luck with missing shadow shifts and stuff lately, and we work through the MI mosaic we sent last time, making sure she knows it like the back of her hand.

I'm going to have her write a simplified version of it herself, when she interrupts me. "I might need to leave a bit later," she says apologetically. "I have a sick cat ...."

Worse than merely sick, as it turns out. Tara's beloved fourteen-year-old Ripley -- the runt of the litter, but as feisty as her namesake from Alien -- is coming to the end of her days. She's got kidney problems, and she's ... she's not well. And Tara's barely keeping it together as the words tumble out of her.

Well, if there's one person around here who's going to understand about that ... I tell her about my own too-recent experiences with Jake and Zenobia, and as gently as I can, I encourage her to go home and spend some time with Ripley. I don't remember much about what I was doing at work the weeks they died, but I remember clearly the time I took off from work to spend with them toward the end. It's like gold in my hand.

"I don't know," she says, "your time is so valuable ...."

"My time is valuable," I agree quietly. "Right now, Ripley's time is even more valuable."

Tara goes home.[1]

[1] This was a hard post to revisit, having just recently lost another cat (Indiana). Life goes on, though, you know: Tara recently got another cat, whom she adores. And I know I will, too. Not yet, though. Not yet.


Spirit Sol 871

We're continuing the extended RAT-brush campaign on Progress, but it hasn't been going well. For one thing, the RAT stalled again, possibly due to a pebble getting jammed in the works.

At first there's talk of blowing off the IDD work for the sol and just doing remote sensing, but cooler heads prevail. We end up doing an MI of the RAT hole, as much for engineering diagnostics as for science reasons -- if there's a pebble, we want to spot it. Then we'll image the RAT with the PCAM and FHAZ, so that if the putative pebble's lodged in the machinery, we'll have a shot at spotting it.

All in all, reasonably routine. The one thing that makes me nervous about this sequence is how close we're getting to the soil. As the RAT hole gets deeper and deeper, we move the MI in closer and closer. But since the MI mosaic that images the RAT hole also moves the MI over portions of the soil outside the RAT hole, we get closer and closer to the soil there as we go.

Our minimum standoff is normally about 17mm, but thisol it's going to be 9.5mm. Later we get updated numbers and find that it's not quite as bad as that, but it's close, at 9.85mm. That's within our 1cm error budget for the instrument positioning system, and I can't help disliking it. If you'd worked with Angry Bob[1], you wouldn't care for this, either. Imaging from the last time, when we were only about 1.5mm higher, shows that we should have acceptable clearance, so ... well ... it's one thing to be nervous, it's another to be unreasonable. I shrug and send the commands.


[1] The nickname for original rover driver Bob Bonitz. If you're curious, here's the tale of his nickname (see that post's footnote).


Spirit Sol 865

We're done. Or as near as damn it, anyway. We pretty much just have to pick up the sol A-857 sequence -- the one that did the last brush on our soil target, Progress -- and resend it. As usual, it turns out not to be quite that simple, but it's not much more complex, either. All around, an unusually easy sol.

So when Nicole Spanovich alerts me, I'm able to find time to remotely tell my TiVo to record Steve Squyres on the Stephen Colbert show tonight. (Gotta love that online scheduling. Magic!)

And Matt has started exploring RoSE macros.[1] He cranks out one to do the IDD safety-deactivate sequences, progressively making it more and more elaborate.

Ashley looks at him with mock pity. "He's got macro fever," she mourns.

Good. Our macros are getting a bit long in the tooth, and someone needs to work on 'em. Matt would be a great choice.

If I'm really lucky, macro fever will turn out to be contagious.

[Next post: sol 871, June 15.]


[1] RoSE -- the Rover Sequence Editor, the part of our rover-commanding toolkit (RSVP) that I wrote -- has a feature called "macros," which let you kick out lots of commands for the rover with just a few clicks. I almost didn't think anyone was going to use this feature, but it turned out to be extraordinarily useful for the rover drivers in particular. Huge swaths of our IDD sequences were semi-automated through the use of macros, and later our drive sequences would be as well.


Spirit Sol 863

Spirit's easy. No RPs thisol.

Spirit's interesting, too: apparently, a couple of rocks sitting very near us turn out to have a remarkably high iron content. Word on the street is, the most logical explanation is that they're iron meteorites.

Well, Opportunity found the first ever meteorite on another planet -- SpongeBob, back near the heat shield. But it's still a feather in Spirit's cap. One of many.

[Next post: sol 865, June 9.]


Opportunity Sol 836 (Spirit Sol 857)

Jeng has to leave early for personal reasons. He was RP-1, I was RP-2, and Matt was shadowing, so I step in as RP-1 and Matt as RP-2. It's like a well-oiled machine.

This weekend, we had a little excitement on Opportunity: she got herself embedded in another ripple.[1] It's just like it was back at Purgatory, only not as bad: having learned our lesson from Purgatory, we had a slip check that stopped us before things got too hairy. Nevertheless, we're jammed in there pretty well, and it's going to take a bit of work to get ourselves out.

Today, we're just starting that process, commanding 5m of driving straight ahead. We'll take high-quality before-and-after pictures, along with a whole bunch of lower-res images in between. If we're lucky, we'll make about 1cm of progress. But it'll be just about the best-documented 1cm drive ever.

[Next post: sol 863, June 7.]

Courtesy NASA/JPL-Caltech. Aw, crap. Not this again.


[1] This ripple was later named "Jammerbugt," after a Danish municipality, for some reason known only to Steve Squyres. I had recently been promoted to the rover driver team lead, and one of my first responsibilities was to lead the anomaly investigation into the Jammerbugt embedding.

I had lots of fun with this assignment -- my final slide included a picture of Captain Kirk ("Risk ... is our business") alongside a quote from Jim Erickson ("What I don't want to see is that we lose our nerve"). But more importantly, the investigation turned up something called the "Sullivan Asymmetry," which we named for Rob Sullivan, the science team member who noticed it and told us about it.

It turns out that the ripples were softer on one side and harder on the other, because as the Martian wind blew across them, almost always in the same direction, it deposited softer material on the leading side and harder material on the trailing side. This knowledge would help us avoid future embedding events because it meant that when in doubt, there was always a harder-packed side we could swerve toward. While that wouldn't have prevented this particular embedding (the root cause was a miscommunication between teams, and we solved that in a different way), it probably did prevent lots of future embeddings.

Oh, and one more thing about Jammerbugt. Rover driver Paolo Bellutta called it "Britney Spears Ripple" -- "Oops, I Did It Again." Funny as hell, but I decreed that we are not naming anything on Mars for that person, and that was that.


Spirit Sol 852

Retract the APXS and swing the IDD away for imaging, then return and place the MB where the APXS was. Simple enough; we could do it in our sleep. But this is the third try, the first two having been frustrated by random uplink failures -- nobody's fault either time, just plain bad luck.

Theoretically, we just have to roll the sequence forward to thisol and deliver it, and we're done. But Chris and I can't help thinking up better ways to do it, and in the end we pretty much rewrite the whole thing.

Al Herrera notices the activity and perks up. "You guys just changing the sol number?"

"Nope," I admit, "we had five minutes, and thought of something else."

"Not surprised," he chuckles, and returns his attention to his laptop.

Well, the new way is better, it really is. Can I help it if Chris and I are perfectionists?

[Next post: sol 857 (Opportunity sol 836), June 1.]


Spirit Sol 842

Meanwhile, back on the other side of the planet ... Spirit's been working on this lengthy IDD campaign on the soil in front of her. The other day, they tried to RAT-brush the soil (yes, really), but instead of the RAT hovering over the soil as they brushed, they apparently actually made contact.

We don't like it when stuff like that happens: the ground should be where we think it is, damn it. So today we're doing a bit of an engineering experiment, touching the soil, putting the RAT into the hover position, taking lots of images as we go. Science-free.

Since the mission started, there's been a bit of a friendly rivalry between the two rovers, particularly when it comes to total odometry. At times, Spirit's been ahead, and at other times, Opportunity's been out in front. Now, with Spirit not driving at all for six months -- and probably never driving long distances again -- Opportunity is poised to take the crown for good. As if to rub this in, Brian Cooper came downstairs and taped to the rover drivers' workstation a fortune cookie fortune he got just the other day:


[Next post: sol 852, May 27.]


Opportunity Sol 814 (Spirit Sol 835)

When I walk into the SOWG room, I'm startled to see Steve Squyres sitting there. Turns out he's out here so he and John Callas can meet with the NASA bigwigs about another round of funding for the mission. I wish him well, of course, and I ask, "Is it going to be harder now that Spirit's down a wheel?"

"The Spirit site is awesome!" Steve exclaims. "There's a whole class of science you can only do sitting in one place, and we've never done it at the Gusev site. No, no, we've got a great story to tell. It's just that there's only so much money to go around." And they might cut one of the rovers in order to save that money. Ulp.

He also tells me Rob Sullivan was thrilled with the MIs we took yesterday. "Rob was really, really psyched when those MIs came down -- you shoulda been there." Which is always nice to hear. Rob's a super-nice guy, and I like it when he's happy with the science we get for him.

The SOWG is over in 13 minutes -- shorter than an engineering keepout, as somebody points out -- and we're off. As we're wrapping up, everyone takes a moment to wish Steve good luck.

But Brenda's not worried. "If Steve Squyres can't get people excited about Mars, then the world is coming to an end."

"That's not the issue," Steve says modestly. "The blood's running ankle-deep at headquarters."

"Yeah," Matt Keuneke chimes in, "they'll be like, 'No funding unless you find a dinosaur bone.' And then they'll probably ask, 'How big is the bone?'"

Well, that's reassuring.

After the usual back-and-forth with Tim Parker and Matt Golombek, Paolo and I plan a 52m drive, including the autonav chunk at the end. Quite respectable progress, if it works out.

As we're wrapping up, Steve returns from the long meeting with the money men. "I went in thinking they're thinking the easy thing to do is to cut Spirit. So I hit Spirit so hard, one of the scientists asked whether we should shut down Opportunity." He grins wickedly. "Then I showed 'em Victoria Crater."

He sighs. "I expect there will be belt-tightening and cuts, but I expect we will have two rovers for the next year and a half."

That funding will have to come from somewhere, and it might end up coming from one of the other Mars missions. Steve offers his opinions on where that cut should be, should it absolutely have to be made. "If it were up to me and I had to cut one, I'd cut the U.S. participation in Mars Express. MEX can live without us. Then I'd cut MGS. MGS is not that great, with just the one camera working -- Odyssey can do everything MGS can, plus -- but MGS is doing great science and MOC is a great camera, and it would be a shame to lose any of it."

[Next post: sol 842, May 17.]


Opportunity Sol 811 (Spirit Sol 832)

At the end of the SOWG meeting, I point out to the whole team what I was able to mention to only a few people yesterday: that we're now up to sol 810 on Opportunity, fully nine times our nominal mission. And there is much rejoicing -- applause and cheers on the audio net.

The drive went beautifully, just perfect. We're perched on the side of the ripple just as we'd hoped. Only problem: there's no visible banding.

Well, it's barely visible if you stretch the images juuust right. But apparently this is something that disappears up close; it's a lot more visible if you look off to either side, along the ripple. That's a weird effect. It's plainly some sort of optical illusion, but I have no idea what causes it.

The bad thing about being perched on the ripple like this is that we took an energy hit, our available energy dropping from somewhere around 440 W-hr to about 386 W-hr, due, apparently, to the 10 degrees of unfavorable tilt we picked up.

Anyway, we're here, and we're going to IDD this thing. While Jeng plans the drive -- a surprisingly aggressive 40 meters -- I work out the IDD sequence. It's unusual for an IDD sequence, a 1x10 mosaic -- basically a vertical column of images all the way down the face. Since the local ripple surface is so uniform, the sequence ends up being nice and clean; I'm quite proud of it. And I hope it works.

[Next post: sol 835 (Opportunity sol 814), May 9.]


Opportunity Sol 810 (Spirit Sol 831)

I wasn't on shift yestersol, Paolo and Matt were, but they called in me and Jeng to give them some advice on driving in an area where we lacked good imaging. Our advice was to suck it up and use autonav. They did, reluctantly -- but it turned out that autonav got us within about 50cm of where we wanted to go, plenty close enough, and we're ready for more.

(Incidentally, when was the first time we had two RPs on shift neither of whom was from the original set? (I tend to think of Jeng and Ashitey as originals, though strictly speaking they weren't.) It wasn't yesterday, surely, but I don't know when it was. Ah, I'm such an old-timer now.)

But we're not going to charge down the trench, going for distance. Instead, we're going to try to address a question that the scientists have been kicking around for a while: what causes the banding we're seeing on these ripples? It looks as though there are several sets of ripples, one atop another, with slightly different albedos -- and what would cause that?

To try to find out, we're going to drive up on one and plant the MI on it. That's a slightly scary and just plain odd thing to do; we normally try to avoid these ripples, not climb them. But it's all for science, you know. So it's game on.

I'm RP-1, and Matt's shadowing me today. Since he's come up to speed on the driving (so to speak) impressively fast, I decide to hand him the reins and see how he does as RP-1. The short version is that he does great, planning a very nice approach where we scoot about 17m down the trench, then turn and charge right up onto the ripple. There's a bit of concern about whether we're going to hit the ripple in our post-drive IDD deploy, but then we realize that the ripple's only about 21cm tall, so the IDD would stay completely above it even if we were buried up to our wheels in the thing.

To make things more complex, though, we have very strict limits on our comm headings. If we're about 3 degrees off one way, we'll get a paltry 23 kilobits -- that's kilobits -- and if we're about 3 degrees off the other way, we'll be at a heading where the PMA will prevent us from talking to Earth at all, so we'll get nothing. Fortunately, ripples are broad, and we should be able to hit it spot on.

Though outwardly cool, Matt confesses he's not completely confident -- "I'm not gonna sleep tonight," as he puts it -- and while that's a feeling I know well, I have a lot of optimism about this drive. As long as we don't end up stuck in the ripple, and we've taken good precautions against that, it should be like hitting the broad side of a barn.

Should be. At least, that's what I'll tell myself when I'm trying to sleep tonight.

It's not until we've wrapped up for the sol, and I'm back in the office talking to someone, that I realize we overlooked something important. I literally run back upstairs to the Opportunity sequencing room, hoping other people will still be there.

Luckily for me, there are several people who haven't left yet, so there's someone to tell. "We just planned sol 810!" I exclaim. "That's nine times the nominal mission! How did we miss that?"


Opportunity Sol 807 (Spirit Sol 828)

We don't have much drive time thisol, so it's going to be a short one. We'll retract the IDD, then drive about 30m across one ripple and up to the next (almost nonexistent) one before we'll more or less run out of time.

And that's how we'll start down the Goodnight-Loving trail. No, I am not making that name up: it's the actual name of an Old West cattle trail, in keeping with our practice of naming our Victoria-bound drives in that fashion. But I can't help getting the adolescent giggles over this name, and I make a couple of jokes about it in the uplink report. ("And finally, on behalf of all the rover planners, I'd like to welcome all of y'all to the Good Night Lovin' trail. Oh, yeah." It's funnier when you read it like Barry White.)

[Next post: sol 831 (Opportunity sol 810), May 5.]


Opportunity Sol 804 (Spirit Sol 825)

I think this is the first time I've done IDD work on Opportunity -- other than the post-drive stow and unstow, that is -- for something like six months, when she had the IDD anomaly to start with. Not much IDD work has been done at all, in fact, and I'm somewhat out of the loop. And today's not just a simple tool placement, it's the whole MI/RAT/MB/APXS/MI/MB campaign over three sols, and I'm RP-1.

So I load up the sequences from the previous time we did IDD work, a couple of weeks ago, and start looking through them. They have a couple of features that surprise me. Last I heard, we were keeping the joint-1 rotor resistance to 58 ohms unless we had a fault; these sequences are using 80 ohms for almost everything except the RAT placement, which uses 75. And last I heard, the shoulder azimuth joint was supposed to stay in a range of about five or 10 degrees from the center so that the arm would be relatively decently positioned in case we lost the shoulder-az joint entirely, but these go to about 25 degrees from the center.

Clearly, I'm not up to speed here. Paolo (who's RP-2 today and is as confused as I) has the right suggestion: go ask Ashitey.

We do that, and Ashitey puts it to us like this: they're working on the extended-extended-extended-extended (or whatever we're up to now) mission proposal, and they want to include a picture of Victoria Crater in it. They don't want to blow a whole lot of sols on the way there, so they've gone ahead and raised the joint-1 rotor resistance limit to ensure that the IDD does not fault out.

I understand this, and of course I'd like to get renewed funding. But, as I point out to Ashitey, if we destroy the arm because we set the resistance value inappropriately high, we'll lose a whole lot of sols. He grins -- it's clear he had the same argument and lost it, and if he lost it, I have no hope. But I decide to try anyway, which means talking to Jake Matijevic.

When I talk to Jake, I decide I really just want one question answered. I put it to him bluntly: "Is this a safety risk to the hardware?"

He shrugs. "We probably don't really need to raise the rotor resistance from 58 ohms at all; if anything, we've seen a possible indication of some recoupling in the motor windings." That would mean we'd need only the nominal amount of current; we could reset the rotor resistance back to its nominal value of 29 ohms and the IDD would work fine -- at least, while the winding was connected. "But is there a risk to the vehicle? No, not really."

He also says it's fine to move the azimuth out of the 5-10-degree band. ("Whatever science wants," are his exact words.) I have enormous respect for Jake and for his knowledge of the spacecraft, and if that's what he tells me, I'm going to believe it. So we go ahead with that. I'm keeping my fingers crossed, just the same, though.

The sequencing itself is a bit tricky but not terribly bad. Some of it I can copy from the stuff they did two weeks ago, which helps. I'm surprised to find that I'm a little rusty, but it all starts coming back to me.

Since this is a multi-sol plan -- 804, 805, and 806 -- we have to have our usual worry about what happens if one (or more) of the sols' IDD sequences is not activated. Obviously, there's no risk to Opportunity's safety if the 806's IDD sequence doesn't make it. If 805's sequence doesn't make it, we'll be OK, since 806 starts with a tool-change command that will do the right thing whether 805's sequence ran or not.

But if 804's sequence doesn't make it ... huh. We've always been protected against the first sol's sequence's not running by the fact that we need to unstow on that sol; if we're not unstowed, the other IDD commands are smart enough not to do anything at all. But since we're unstowing at the end of every Opportunity drive now, there's no such protection any more.

So Paolo and I have to take a careful look at it. After a great deal of to-and-fro-ing, we work out that we're just fine as it stands. The IDD is at the ready position, with the turret held just in front of the vehicle, APXS pointing forward and slightly down. Both sols 805 and 806 begin with a tool change command that will first retract the active tool -- the APXS, as it happens -- 13cm. But in its current configuration, if the rover tries to retract the IDD 13cm, the IDD would collide with the body. The rover's smart enough to know this, and it would refuse to execute that or any subsequent IDD motion command until the fault was explicitly cleared from Earth.

Thus, we're fine, and we can rest easy. With our fingers crossed.

[Next post: sol 828 (Opportunity sol 807), May 2.]


Opportunity Sol 800 (Spirit Sol 822)

It's sol EIGHT FREAKING HUNDRED on Opportunity today. (Somewhere in there, by the way, they used the target name "Fort Scott" -- a stop on one of the Old West trails -- and I missed it! Darn!) Eight hundred sols, and we're still driving -- and driving, and driving. We've just come out of restricted sols, and the plan for at least the next four sols is to drive every sol.

Which is just fine with me and Jeng. So we plan your basic 50m drive, composed of about 30m of eastward traverse (including two ripple crossings) and then a southward segment with autonav.

We also spend some time discussing Jeng's "RP Lessons Learned" assignment. This was a terrific idea by Chris Leger to identify a bunch of past sols where we'd screwed something up, and set the sequences in front of the newer RPs (each of them paired with a more experienced RP) and make them figure out what we'd done wrong. And, for extra added bonus evil, they then have to present what they learned to the whole IST team. The idea is to increase their general skill level, as well as making them more paranoid. It seems to be working.

Jeng's not really a new RP -- not like Ashley or Matt -- but he got his share of sols anyway, and I was paired with him. And one of his sols was the Mazatzal approach.

Oh, yes, the Mazatzal approach. As if I could forget. I've been feeling terrible about it for two years now.

I feel a little better when I see Jeng struggling to figure out what's wrong with it. It's not schadenfreude, it's just that I realize now how tough it actually was. The problems are subtle ones. First, we misjudged the amount of slip we were likely to see. Second, we aimed for a position where we'd just be able to reach the Mazatzal features of interest -- but since we knew we'd slip, we should have commanded the rover to go farther, so that we'd tend to slip toward the edge of reachability but maybe not out of it.

Just seeing that Jeng is having this much trouble -- and he's got a lot of experience now -- starts to heal that old wound a bit. And when we talk about it and he sees the problems, he makes me feel even better. "That was tough!" he exclaims. "And back then, we couldn't use visodom because of that bogus update on Opportunity. Without a way for the rover to correct for slip, how were you supposed to get this?" He shakes his head. "You should give yourself more credit."

It's nice to realize that in two years, I have learned to give myself more credit. And I think I'll take his advice and practice the skill now. That won't go into the presentation, but it might be the most important lesson this RP has learned.

[Next post: sol 825 (Opportunity sol 804), April 29.]


Spirit Sol 821

John Wright and I show up to drive Spirit today, but she's just taking pictures. No IDD, and -- just as will be the case for most of the next six months or so -- no driving.

So we leave.


Spirit Sol 817

It's a slow day. We're planning two sols of IDD work today -- 817 and 819 -- but they couldn't be much simpler. One's just a tool change from the APXS to the MB, and the other retracts the MB and swings it in so that it's out of the PCAM FOV.

Plus, I already wrote them, a couple of days ago. So all that remains is for me to turn this into a training opportunity, watching Terry laboriously rewrite them, from scratch.

John Callas alerts me that we have visitors -- a passel of Indonesian legislators in the tow of Congressman Dreyer -- but since they don't speak much English and are foreign nationals from a "designated country" anyway, they don't come by the sequencing area. I never even see them.

I'm not really sure they were ever there. Maybe it was just some kind of practical joke Callas was playing on me, and it went over my head.

[Next post: sol 821, April 25.]


Spirit Sol 811

The LTP guy is driving Alicia Vaughan crazy. His slides abbreviate "PANCAM" as "PC," and he pronounces "ODY" -- Odyssey -- as "oddy." She giggles and rolls her eyes when he does this. "New guy," she groans.

We don't have much to do today, just a tool change from APXS to MB, and Ashitey and I decide to make Terry (shadowing us today) do it. So we have even less to do.

But we do have a Spirit Drive Meeting to go to. Our goal is to figure out what we can, and should, do about driving during the winter.

It very rapidly starts to look like the answer is: nothing. We need to gain 5-6 degrees of tilt in order to get a 15W-h power increase. It's doubtful we could do that, and anyway we think we don't need to: the current 10.7-degree tilt is believed to be survivable, though not with much room to spare.

Unfortunately, the current IDD work volume is "not very exciting," as Squyres laments. There are layered rocks and vesicular basalt nearby, either of which would be better -- indeed, we could reach the layered stuff just by turning clockwise, something we definitely can do. And it's appealing on other levels: we'd be facing south, with the solar panels aimed north, which is gives us less shadowing, is better for science, and improves our HGA comm. The problem is that our UHF comm will be poor, and since we relay heavily on that -- even more during the power-poor winter months -- that consideration will probably trump all the others.

The rules we decide to adopt are that any drive must:

  • Be in the best interest of survival.

  • Improve comm for the winter.

  • Improve our solar tilt.

  • Improve science.

  • Include a contingency plan to return to safety in case the drive doesn't work out.

We might be willing to relax these constraints, but they're pretty discouraging when you look at them. There are a lot of ways to make things worse, and not many ways to make them better. So we're likely just to sit here, take lots of pretty pictures, and do a little IDD work. And wait for spring.

At least Spirit's IDD is working in the meantime. I'd forgotten about this until Ashley reminds me, but one of the possible explanations for our poor rover's lame wheel was that some circuitry had failed on board -- and it so happens that one of the possible failures was one that would have taken the IDD with it.

See what I mean? There are lots of ways things could be worse.

[Next post: sol 817, April 21.]


Opportunity Sol 789 (Spirit Sol 809)

Results from the previous drive were ... well ... suboptimal. And confusing, to boot. Opportunity detected excessive slip when climbing the first ripple, so we only made a few meters of progress. The confusing part is, we set her max allowed slip to 70%, and none of her reported slip numbers appeared to exceed that value. So we sort of know why she stopped, but not exactly.

While Matt looks into that, I survey the current state of the rover. The maximum reported slip numbers were still higher than they should have been, and we want to know whether it's safe to continue. Flying around for a while in RSVP, plus looking at the telemetry, clears things up nicely: we're just barely over the crest. In fact, when I look carefully at the rear HAZCAM and brighten the shadowed areas way up, I can just make out the crest of the ripple running right in front of our rear wheels. Another few cm and we'll be completely on the downhill side.

Foo. Well, the four wheels we can see aren't dug in, and the slip check turns out to have failed because of buggy -- well, let's say surprising -- behavior in the flight software that makes visodom slip checks generally more conservative than we thought they'd be. We told it to stop at 70% slip, but it was actually using a threshold of something like 52% when it stopped. So we're safe to continue, and that's what we do.

We don't go quite the same path as we'd picked out previously, though. It occurs to me today that our plan was to drive about 20m through small, but soft, transverse ripples, then hop a regular old ripple at a not particularly wide saddle point. In other words, after about 20m of picking up an unknown amount of slip, we're going to aim for a small target. And if we miss, we could have another Purgatory on our hands.

When I put it to myself in those terms, I can see what a bad idea it was all along. We could use visodom to compensate, but then we'd run out of time before we even got to the ripple, so that's pointless. Instead, I shoot for a few meters farther down the trough, where I can put a juicy patch of outcrop right in the IDD work volume. That's what Steve wants anyhow -- a little outcrop to IDD for the weekend -- and since it's the safest thing to do anyway, I'm happy to oblige.

[Next post: sol 811, April 15.]


Opportunity Sol 787 (Spirit Sol 807)

Autonav worked, and it worked great. We came this close to making it all the way to the autonav waypoint, putting a total of 58.97m on the wheels. More important than the extra dozen meters or so, of course, is that it's back in everyone's consciousness. Also, now we can truthfully say we've proved it works, and we can use it again.

But today won't be the day. Today's drive is a fairly complex one, involving not one but two ripple crossings. We hop a nearby ripple, zoom down the trough for about 25m, and then hop another ripple onto a patch of outcrop at the far end-- total, about 35m. Jeng's RP-1, and he basically works out the whole thing.

Which leaves me with just one thing to do: take the credit.

I mean, assuming it works.

[Next post: sol 809 (Opportunity sol 789), April 13.]


Opportunity Sol 785 (Spirit Sol 805)

Our previous drive went splendidly, gaining us about 57m toward Vickie (my little nickname for Victoria Crater). This leaves us an estimated 1603m from Victoria. During the SOWG meeting, I do the math, and am able to announce a milestone. A literal one, at that: "We're about 1603m from Victoria Crater, and a mile is 1609m -- so we're less than a mile away!"

Applause, applause.

(Speaking of the SOWG meeting, our SOWG chair today is none other than Steve Squyres himself. The last time Squyres was SOWG chair on this rover was somewhere in the 400s, back when we were in Endurance. I will try not to interpret this as some kind of hint that he lacks faith in Spirit.)

The big question of the day is just how far we're going to go. The next trough over (to the east) is more appealing than the one we're in, so we'll hop into it. Well along it -- 45m from our starting point -- is a small patch of outcrop. We can reach that without much trouble, and we can do so with considerable time to spare.

Then what? The most appealing option is another patch of outcrop 20m beyond that (65m from here), but I don't think we can reach it. There's a big mound -- a transverse ripple, I think -- just past the 45m outcrop. It's too big for us to go over and wide enough that we don't have room to go around.

Off to the other side -- to the east -- is another patch of outcrop. This one's only 58m from our starting point, so it's less appealing than the 65m patch (though it does let us go a bit eastward, which is something we try to do when we can, since Vickie's a bit east of south). However, between us and it is a bit of a step, or a slope.

The likely reason for that, judging by what we've always seen in this terrain, is simply that it's a perfectly safe, shallow slope that we can't quite see from here. However, we can't prove it's not a dangerous step.

Now, as it happens, we have a tool in our toolbox for just such occasions -- autonav! We haven't used autonav on this rover much since Purgatory. Frank and I had just reintroduced it and used it for a few drives, when Opportunity's IDD started failing. Since then, it hasn't been used at all.

But Frank and I are on shift together today, and, well ... autonav it is!

I honestly don't care whether we make a whole lot of distance on the autonav segment. What really matters is that we will have done it, and can start incorporating it into Opportunity's drives again. If she even takes a step or two on autonav, that will be a roaring success.

I also introduce another innovation today. For a long time, we've wanted RoSE to indent IF statements in our sequences, so that you could see the sequence structure more clearly as you wrote and walked through them. I got around to writing this code at last, figured out a way to backport it from the development branch to the installed branch, and took advantage of a scheme I worked out some time ago to dynamically apply the patch to the running (installed) version. I do that for the first time today, and wait to see how long it takes Frank to notice.

It takes him about, oh, I'd say, all of five seconds. Then we get another idea: we decide to have him running with that patch during the walkthrough, and we'll see if anyone notices.

Frank loves this idea, so we do it. All I can think about during the walkthrough is when somebody will notice -- but nobody says anything. I'm quite disappointed. But then, as Frank's wrapping up, Saina says, "Is this the first day using the new RoSE?" From the reactions around the room, it's clear she's not the only one who noticed. (But, amusingly, Julie didn't, and Frank comically refuses to point it out to her. "You've got to see it for yourself," he shrugs.)

"Yeah, this is the first time we've used it."

"Is it tested?" she asks.

In unison, Frank and I answer, "It is now!"

"Back when Jim was mission manager, you think he woulda let us get away with that?" Frank asks me rhetorically. "Times have changed."

They sure have. Back then, we wouldn't have tried to get away with that.

[Next post: sol 807 (Opportunity Sol 787), April 11.]


Opportunity Sol 782 (Spirit Sol 803)

Saina's leaving the project. That sucks so bad, I can't even tell you.

I can't blame her; she's a mechanical engineer, and she wants to build stuff. But she's so great at ops, I'll really miss her.

"I think they want me to build the cap part of the aeroshell," she says.

I point out that this means she'll be designing the part of the spacecraft that gets to Mars first, and her face lights up. No way she's coming back to MER ops now.


Today's my first shift driving with Matt Heverly since he started working solo RP-2 shifts. And we've got an interesting one. The end of our current trough is in sight, and as we stare at the images, we start to realize there's nowhere we can safely hop to the next ripple between here and the end.

Unless ....

When I look down at the nearby edge of the images, just off our left side, it looks like the ripple sort of curves out, like an opened gate. A review of our HAZCAMs and the previous sol's images confirms this; there's a low spot almost dead in front of us. We just have to scoot a bit forward, then back across the ripple at a low point and we can head down our neighbor trough just dandy.

So that, of course, is what we do.

As of the start of today's drive, we're about 1660m from Victoria Crater. Put another way, that's only two Victoria-Crater diameters from Victoria Crater itself. And counting.

[Next post: sol 805 (Opportunity sol 785), April 9.]


Opportunity Sol 780 (Spirit Sol 801)

Opportunity's schedule will be late for the rest of the week; we're driving every day, and we're not starting until noon or later. So I'll pretty much be in heaven. Since we're devoted to driving this week, we're keeping the science lean; next week we enter restricted sols and the scientists can have their turn again. But for now, the SOWG meetings are short and sweet. This one is over in something like 20 minutes -- not a record, but not too far off.

Downstairs, our sister has good news: Spirit's dragged herself out of the muck. She followed her tracks back out, and while the HAZCAMs show a bit of the white stuff near her new tracks, they also show she's back to making good progress.

Naturally, the next question is, now what? Word is that they've given up on McCool and have decided instead to head for a ridge to the south, which we passed coming in. It's not as far away as Home Plate (or McCool) and not as scientifically interesting as McCool, but we think we can get there, and it has slopes we think we can survive on.

So that's good.

But according to Brenda, the news is not altogether rosy. "John [Wright] thinks we're seeing the last and final days of Spirit," she mourns. "I don't think I want him driving my rover if that's the case."

Well, maybe I'll have to give him a nice pep talk. Meanwhile, I've got my own rover to drive, or anyway, to watch Jeng drive. Thisol's drive couldn't be much more boring -- we could practically do it with a single waypoint -- but I don't mind that now and then.

For one thing, it gives me time to slip out for an interview. Some British guy named Barney is doing a piece for a British engineering association -- I gather it's something like the ACM or the IEEE. They're concerned that British kids aren't sufficiently interested in science and engineering, and they want him to make a video to get the kids hyped up about it.

So I meet him in the von Karman museum, and I get there early enough to watch him do the interview before mine. It's with a very nice and articulate Brazilian scientist we have working here -- Rosalie something, I think, is her name; I've never met her before. If I didn't know better, I'd swear the interviewer guy was kind of hitting on her afterward.

Anyway, then it's my turn. The guy asks me a couple of warmup questions while his two-man crew -- they're brothers -- rearrange the lights so they can shoot me with the rover model in the background. The interview itself seems to go really, really fast, and I don't think I'm at my best, but both he and Natalie (the press office lady who arranged to get me for the interview) are really positive about it, so who knows. I'll see if I can get them to send me a DVD of it, and then maybe I can evaluate it for myself.

I want to stay and watch the presentation that's just starting next door, in the auditorium. It's a couple of teachers from New Hampshire who have worked the rovers into their curriculum. They have their students build LEGO (Mindstorms?) rovers and drive them around in a simulated Martian landscape. I had some great teachers, don't get me wrong, but why couldn't I have had a couple like that?

But I don't have time, so with unaccustomed regret, I trudge back over to the sequencing room. (Okay, it's not that bad!) I'm there for about half an hour when Saina says something about wanting to go to the LEGO talk later this afternoon.

"Later this afternoon? No, it's going on now -- you're missing it," I tell her.

"What?!" she exclaims.

A couple of minutes later, we're all heading over there together to watch what's left of their presentation. It helps to have a mission manager who likes LEGO.

The bad news is, we miss the LEGO part of the presentation. The good news is, they give us a copy -- and accept our invitation to come by the sequencing room and see the real thing. So I give them my usual demo -- which I do flawlessly, best one I've ever done -- and Jeng shows them the cool 3-D glasses, and they're digging us and we're digging them, and it's all a big group hug thing. But in a good way.

[Next post: sol 803 (Opportunity sol 782), April 6.]