Spirit Sol 52

First thing on my agenda -- literally, just a few minutes after waking up -- is the WBBB interview. I've begun preparing for interviews by writing down a list of points I can make when the opportunity arises. (This is something our media trainers suggested, but I had to discover the truth of it for myself -- it's my way.) I make the list before calling in, but turn out to need very little of it, which is fine. It's a short interview anyway, just five or ten minutes. And it goes better than last time, as they don't ask me about any sporting events I haven't heard of, like, you know, the Super Bowl. As usual, they're very nice to me, and reasonably fun to talk to.

Just as I arrived at the same moment as Chris Leger yesterday, I run into Alex Hayes today. Alex is one of the scientists -- one of the younger ones, a student of Squyres who's just gotten his Master's. He's going back for his Ph.D. after the mission, and with something like this in his background, I imagine he'll get in anywhere he wants. Alex is a classic New York or New Jersey native, loud and brash. I usually find this personality type difficult to deal with, and Alex was no exception at first, but I've come to like him.

Our conversation is interrupted by a cell phone call from Mark Maimone, who has good and bad news about the drive. The rover got nearly to its destination but, near the end, was spooked by a slight downslope it saw in front of it. The downslope is not really a hazard, but the rover thinks it is, and that problem has been known for a long time -- indeed, Mark has long had a software patch to fix it, but they don't want to upload it yet. Today that decision bit us, but not too badly. The rover was about 3m from its target when it stumbled onto the slope, decided it was a hazard, and wasted the rest of the drive time making short back-and-forth staggering motions, hunting a way around the perceived obstacle. There wasn't one. We still ended up near the drive target, and it was a record-setting distance in terms of wheel odometry, though my record still stands (for now) if you measure start-to-finish distance covered.

The news is actually a little worse than that, as Rob Manning reports at the downlink assessment meeting. (Incidentally, if I ever have to get bad news, I want Rob to be the one to deliver it: he's so inoffensive and funny. This isn't the first time I've seen him manage these situations, and as before, he leaves the scientists more amused than angry.) The rover's downlink data rates are sensitive to its final orientation, so we try to tell it to point in a good direction after the drive. This is important enough that we even check for time-of-day limits on the drive and clear them to give the rover a few extra minutes to establish the correct orientation. But when we wrote the commands to do that, we failed to consider the case where an individual waypoint command times out, only the case where the global time-of-day limit is reached. As a result, the error wasn't cleared, and the rover ended up facing in a suboptimal direction. We still got the comm pass, but we got a lot less data than we wanted.

Since it was his drive, Chris -- as much a perfectionist as I am, and that's saying something -- is very upset about this result, as I would be in his shoes. So I try to cheer him up, pointing out that we were within our 10% error budget (the scientists are just spoiled -- they've gotten accustomed to our perfection, I guess), and that this will bolster support for Mark's nav-code patch, which only yesterday seemed to have flagged. Chris doesn't look any happier. I stop trying.

The scientists discuss the implications of this outcome. They'll have to generate a little less data tomorrow, for one thing, and because there's more uncertainty than normal about the post-drive conditions, the MTES experimenters aren't sure that their post-drive experiment did what they wanted. (But Ray reminds them of something he's said before: "There's no bad data.")

At least we know what to do tomorrow: we'll finish the drive, this time in blind-driving mode, so the rover won't be spooked by the non-hazard.

The happy news is that we've now driven 182m of the 300m we need to cover for mission success. As it happens, this is measured by wheel odometry, so the staggering at the end of the drive counts, even though it didn't take us anywhere. (Theoretically, we could achieve mission success in this respect by just driving the rover back and forth over the same spot, in one-meter segments, until we racked up 300m on the wheels. Come to think of it, that's not so different from what Opportunity is doing, ha ha.)

Ray praises the scientists on yestersol's plan. Coming out of the SOWG, we were only modestly oversubscribed (a desirable state; it tends to help in making tradeoffs at the subsequent refinement step, because there's a pool of alternatives to choose from when you cut something). Afterward we discovered power problems, but we were able to come up with a useful plan because the science team had done a good job of setting priorities. We're becoming an energy-limited mission, he says, so good prioritization is only going to become more important. John Grant asks if there are any lessons we can learn from Opportunity, then -- they've been energy-limited from the start, thanks to a stuck switch on a heater. The answer is yes: they create flexible, modular plans, so that the pieces can be rearranged as needed when problems arise.

Just as Ray got used to the term "antepenultimate," now we have another: "preantepenultimate." Tomorrow's drive requires a preantepenultimate turn in order to get into the right orientation for communications. There are widespread chuckles every time Ray stumbles through the word, which he appears to have to think about afresh each time he says it.

We're not sure if it will be safe to IDD, and Ray rejects the idea of trying. Tomorrow, we're data-limited, so we should go with a simple plan, and that's what they agree on.

I decide to skip the SOWG to catch up on my work backlog. Art Thompson comes in as I sit down, and asks the question everyone asks me sooner or later: "Don't you have anything better to do with your life?" I think about it for half a second, and say, "Well, no -- is there something better to do with your life?" "Actually, I guess not," he muses. Art is mock-upset because tomorrow's drive will be so short. "They never let me drive!" he thunders. "Yesterday, they do, like, thirty meters. And then today I'm the TUL, and it's, 'Let's just drive a meter.'"

He's at least half-joking, but as to the other half ... well, I know how he feels. At least I got that out of my system somewhat when we did the megadrive. Art hasn't had the pleasure yet. Maybe he'll get a chance soon.

I spend my remaining time writing a Perl script that should save the SIEs many hours of repetitive work, and finish just as John arrives. Considering this to be a more or less perfect opportunity to go home, that's exactly what I do.


Mike Lawrence said...

I might have missed it before but there seemed surprisingly little communication between the Spirit and Opportunity teams. I would have thought that the scientists would have been really interested in results from both rovers so that say the science meetings would have had both groups there. Also that the software patches and improvements you made would have been useful to both teams. Was it just too busy? I love your humour by the way!

Great blog


Scott Maxwell said...

@Mike Lawrence Since the rovers were on opposite sides of Mars, their schedules were offset from each other by about 12 hours. The science meeting for one rover would happen in the middle of the night for the other rover's science team, for example. Switching from one rover to the other therefore required more or less completely inverting your wake/sleep schedule, so it didn't happen much.

Most people, on both the science and engineering teams, were assigned to one rover and stuck to that rover all the way through the nominal mission. I know a few people switched rovers here and there, and I'm sure the two rovers' science teams communicated via email, but there were real limits on what they could do.

Remember, too, that we thought we had only about a 90-sol mission. On the engineering side, we didn't imagine our approaches would have much time to diverge, and on the science side, I'm sure they thought the worst case was that they'd catch up on the science results from the other rover about three months later.

We did start swapping people later on, when we knew we had an extended mission and our return to Earth time made it feasible. Nowadays, I'd say that more than half of the team switches rovers more or less freely.