Worse, I don't even end up sleeping that late. I wake up more than an hour early, half in a panic over yestersol's sequence. I've consistently been making a particular mistake recently, accidentally telling the software to move the IDD to the wrong place -- not putting the instrument at the staging point above the target first, but moving directly to the target instead. This boneheaded mistake shows up in the simulation and so I catch it almost right away, but yestersol we changed the targeting at the last minute and then looked at the results from an unusual perspective. Am I really sure we would have noticed if I did the wrong thing? If not, the arm will contact the rock unexpectedly and the sequence will fault out, and we'll lose a sol recovering.
I argue with myself about this. John sat there and watched me make the changes. (But am I sure he would have noticed if I'd clicked the wrong button one time?) We review the animation before uplinking. (But we skipped one of the reviews yesterday because we were behind schedule, and we showed the animations from the same unusual perspective that John and I were using -- isn't it possible we would have missed it?) John independently reviews the sequences I write; that's the point of an RP-2. (But ... but ... but ....)
God damn it. I get up and log in from home, find and download the MPEG movie John routinely produces (only I can't find today's movie in the usual spot, so I have to generate my own and download it, which takes forever even on DSL, so I shower while it's downloading). I watch the movie and, horribly, see the IDD do exactly the wrong thing. We are so fucked.
Then I play the movie again, a little slower. Turns out I was wrong: the IDD move was fine; it just skipped by too fast the first time. At a lower speed, you can clearly see the IDD doing exactly what it's supposed to. It's fine; I was all freaked out over nothing.
It's not like I haven't noticed the pattern. I just can't help myself.
At work I review my traversability analysis with Chris, to save him some time -- the less time he has to spend looking for the hazards, the more time he has to plan a path around them. "Opportunity doesn't have to deal with this," I muse. "They can just drive."
"Ah, it's not really driving if you don't have to avoid anything," he says, mock-contemptuously. "Anyone can put a brick on the gas pedal."
I'm not into the whole inter-rover rivalry thing[1], but that cracks me up.
I get distracted for the first few minutes of the downlink assessment meeting by a beautiful, huge book published by the American Geophysical Union: "Mars Exploration Rover Mission and Landing Sites." It appears to be a large collection of papers about -- well, pretty much exactly what it says. Way cool.
LTP puts up a slide with a warning that came out of my uplink report from yestersol, where I was explaining that we were uncertain about the exact location of instrument placements because of the mesh problem that caused us to miss Hawaii and end up higher on the rock face (in Alaska, as Ray jokes). This same problem may cause the RAT to end up in a location where it won't be able to do its job.
Ray asks me to elaborate on this, and I've got the mike in my hand, explaining that it's a very low risk but we're just not 100% certain, when Craig Leff pops in. "We've got the HAZCAMs," he says. "The RAT worked great."
"Like I said, it's going to go perfectly," I say, handing Ray the mike back.
The whole room swarms around one of the projection monitors. It looks exactly right: a pentagram of round RAT brush spots, with the sixth and final brush spot right in the middle. "It's a flower!" someone exclaims. "It's so cute!" "It's better than Mickey Mouse, isn't it?" says Steve Ruff.
The scientists are like kids in a candy store, until Ray calls them back to order. "Okay, folks, let's focus," he says. But he can't resist making his own comments. "It's gonna be a good day," he says. ("It's starting off pretty well," Steve agrees.)
I have recovered my honor.[2]
The meeting is as upbeat as that makes it sound. There's applause for the RAT guy when he walks in. The Moessbauer data is good. And we're getting the hell out of here. "I think the sense of this sol is: drive," Ray says. "Drive, get some data, drive, get some data, then some runout science."
I won't be doing the drive; Chris gets to do that. But I'm happy to get out of here. It seems as if we've been stuck at Mazatzal forever, and I want out.
There is, however, a moment of acrimony, unusual for this meeting. Geoff Landis proposes getting PANCAM imagery of the backside of the antenna, as a way of measuring dust accumulation or something. He wants to take it during the drive, when we're heading at 220 degrees. Ray has clearly heard about this idea, and doesn't think much of it -- it seems to annoy him. "What do you want with that data?" he asks sharply. "It's less dusty, so what? You don't have measurements of the scattering properties, so you can't analyze it."
"We have sol-2 data; we'll compare it with that. This might be our last chance to do this. We're not at 220 at noon very often."
They go around and around like that for a couple of minutes, as Ray just gets more annoyed. "It doesn't seem to me like a well-thought-out experiment, I'm sorry."
Jeff Johnson tries to make it better, gently supporting Ray. "There are other dust experiments we're doing -- solar panels and so on -- those should be good enough."
"And I'm not willing to interrupt the drive to do this," says Ray.
Ray seems to think that will end it, but Geoff doesn't give up. "You don't have to interrupt the drive. We can take the image at the end of the drive. And if we're going to do this, we have to do it now: conditions are getting worse, the sun is getting lower in the sky ...."
"I'm not gonna do it, I'm sorry," says Ray. And that's it.
After a little more discussion, the meeting is over. "It looks like we've made this beautiful flower," Ray says, "so let's get our sequences together for the SOWG meeting."
That's my cue to leave. I decide to go home before anything can spoil my day, and almost make it. Yesterday John and I decided not to tell the rover to use extra memory when computing its on-board nav maps, since we weren't driving very far -- just a 90cm backup -- and it wouldn't have made much difference.
Except that it might have made a lot of difference. We both forgot that the rover's on-board parameters have been set to assume that the extra memory is available, and we're supposed to tell it to use that memory for driving every sol we drive. If we don't, the rover can fail to allocate memory, causing a fatal error and a reboot. That didn't happen yesterday -- we got lucky.
Come to think of it, that is good news.
Courtesy NASA/JPL-Caltech. The RAT mosaic, along with the RAT grind spot, on Mazatzal.
RAT mosaic
Footnotes:
[1] Times change. :-)
[2] I was still -- still -- smarting from the failure of the Mazatzal approach drive. Despite my jaunty tone above, I would continue to be upset about it for months. But you'll hear that part of the story later.
No comments:
Post a Comment