We're planning two sols today, and my part could hardly be simpler: just a single tool change from APXS to MB. Less than twenty-five minutes after I arrive on Lab, I'm done for the day.
Or I should be. But I have to stick around and play both my part and Ashitey's, since he's been asked to look into Spirit's RAT anomaly. This happened some weeks ago: the teeth on Spirit's RAT finally wore away (after 15 grindings, five times as many as its design required). While this was disappointing, in itself it was neither surprising nor worrisome. But what was worrisome was what happened afterward: as the bare metal below the worn-away grind head contacted the rock, the IDD lost contact with the rock face and started skipping around, vibrating the entire arm. This went on for an hour and is extremely dangerous -- it could have broken the arm. Luckily, Spirit survived this event, but we're worried about what could happen if Opportunity goes through the same thing. So Ashitey's looking into ways to keep that from happening.
He's got to be down in the testbed thisol, but he promises to come back for the meetings and for this afternoon's end-of-sol science discussion. (That's no problem. There are sols when we need 1.5 RPs, and sols when we need 2.5. We get 2.0 each sol, so I guess we average right. Thisol is more like 0.5, so Ashitey could go hang out at the beach today for all it really mattered.) "I think we're not going to the summit," he says. "At least, that's my feeling."
"We've got to go to the summit," I exclaim.
"I agree," he nods. "We shouldn't let this thing conquer us." He turns to go. "We should convince the scientists we can get to the top in 25 sols." This is a joke: it'll take considerably longer than that, and we both know it. The only problem is, so do the scientists.
Well, since I've got some spare time, I put it to good use. I know approximately what they're going to want tomorrow, so I go ahead and put together a candidate sequence to do it. That should help my fellow RPs tomorrow. Part of this work involves planning for sol 500. Sol 500! In a few more days, we'll have a combined total of 1000 sols of operations on both rovers.
Between this and other catch-up work, I manage to find ways to keep myself busy until the end-of-sol meeting. The first part of this meeting is an hour of pent-up discussion of science results and what they mean. I haven't been able to attend one in a long time, and usually find these fascinating, but last night I barely slept, and I just can't focus. I snap to when the discussion changes to a strategic one.
Squyres puts the central question to the group: "Where do we go next? Home Plate? The summit? Into the Tennessee Valley?" Basically, this means: do we go around Husband Hill, to its top, or turn our back on it and descend to the valley we can see from here?
I have my strongly held opinion, but I keep it to myself a while. The valley gets more or less ruled out, mainly because there's nothing obviously compelling about it and we might not be able to climb its far side if we went there.
To focus the discussion, Steve asks another question: "The dust storm season is coming. Suppose we had only two to three months to use these vehicles; how would we use them?" The science team's answer to this question revolves around characterizing the local structure and stratigraphy -- basically, wandering around more or less where we are. I can't hold my tongue any longer. If we have only two or three months of life left in this rover, it deserves to go to the top of this hill, and I say so. In his heart of heart, that's what Steve wants, too -- this I'm sure of -- but he's got to be able to make a purely scientific case for it. So he evinces a reserved enthusiasm, and shortly the discussion moves on.
I mainly keep quiet, but I can't resist speaking up when someone asks about our chances of surviving the dust storms. "If we're at the summit," I point out, "there's less dust between us and the sun." This gets a big laugh.
It also gets a serious answer from Steve: "The dust storm season is variable. We could be dead in two weeks from a tau of 5, or we could sail right on through with a tau of 1.2. Both things are possible. Our strategy from the start on this vehicle has been to do as much science as we can, as fast as we can."
I admit I've gotten out of the habit of thinking of our babies that way. They've survived so much, they've come so far, I've been thinking nothing would ever stop them.
But I know something will. I don't know what, but something will.
And, damn it, I want to be on the top of that hill before it happens!
[Next post: sol 502, June 1.]
2010-05-27
2010-05-25
Spirit Sol 495
We're just about to enter a period of restricted sols; consequently, the downlink's late -- 12:40 is the nominal start time. When it shows up, everyone's congratulating me on the success of the drive. But they're more sure it was a success than I am. I really wanted this target rock right in front of the front wheels, and it's definitely farther away than that.
As more data shows up, it becomes clear that we're close enough. The rock is a flattish slab, and the lower part -- the side facing us -- is reachable, though the top is not. But as Albert Yen points out, the top is the dustiest part anyway. They're more interested in the relatively dust-free side part, which we can reach. So we won't need to bump.
The whole situation reminds me, oddly enough, of Mazatzal: a relatively short approach to a flattish rock, and we came up just a little bit short of the goal, but close enough to get the job done. This time, unlike Mazatzal, I don't feel lousy about the result. I've changed.
We plan two sols' worth of activities on this as-yet-unnamed rock. Following a now-familiar pattern, on the first sol, we'll do an MI mosaic, then switch to APXS. And on the second sol, we RAT-brush the spot and redo the MI mosaic and APXS. Back when we did this on Mazatzal, I'd have been frantic. Now I'm hardly breaking a sweat.
As the SOWG meeting wraps up, Albert points out that we need target names. "Are we still sticking with the nautical theme?" I ask. "Because I was going to suggest that we call the rock 'Pequod' and do a Moby Dick theme." This suggestion goes over well. When it comes time to pick some target names, I key the mike again: "We should call one of them 'Ishmael,'" I point out to appreciative laughter. The preliminary IDD target gets that name, though later, based on advice from the RAT team, we switch to a different target, Ahab. Well, that's OK: if you're going to be obsessed with a target for a few days, it might as well be a target named for a character who's practically a synonym for obsession.
I finish both sols' IDD sequences and hand them over to Chris. Then I check out the images from yestersol, which include some very cool MIs. The ones we have are slightly out of focus -- not surprising, since we forewent the MB touch in order to preserve the soil -- but we can see that they're going to be awesome when we get the in-focus versions. One looks like a cliff, the others like a miniature city or something.
Just before the CAM I step out to stretch and run into Mark Maimone. He fills me in on Opportunity's progress. They're seeing 99.5% slip: so far they've commanded 48m and have made 27cm of progress. This in itself is more or less what we'd expect, based on the testbed data. But what worries him is that we don't seem to have risen any yet -- it's as if we're just pushing this stuff in front of us all the way down the side of the ripple. He's concerned that we'll end up at the bottom of the ripple and still be buried, in which case we might never get back above the stuff.
On a lighter note, he points out that our expectations have dropped four orders of magnitude. "We used to do 200m sols, now we do 2cm sols," he says. And we're just as happy to get them.
[Next post: sol 497, May 27.]
Courtesy NASA/JPL-Caltech. The view from the end of the drive, with Pequod just in reach.
As more data shows up, it becomes clear that we're close enough. The rock is a flattish slab, and the lower part -- the side facing us -- is reachable, though the top is not. But as Albert Yen points out, the top is the dustiest part anyway. They're more interested in the relatively dust-free side part, which we can reach. So we won't need to bump.
The whole situation reminds me, oddly enough, of Mazatzal: a relatively short approach to a flattish rock, and we came up just a little bit short of the goal, but close enough to get the job done. This time, unlike Mazatzal, I don't feel lousy about the result. I've changed.
We plan two sols' worth of activities on this as-yet-unnamed rock. Following a now-familiar pattern, on the first sol, we'll do an MI mosaic, then switch to APXS. And on the second sol, we RAT-brush the spot and redo the MI mosaic and APXS. Back when we did this on Mazatzal, I'd have been frantic. Now I'm hardly breaking a sweat.
As the SOWG meeting wraps up, Albert points out that we need target names. "Are we still sticking with the nautical theme?" I ask. "Because I was going to suggest that we call the rock 'Pequod' and do a Moby Dick theme." This suggestion goes over well. When it comes time to pick some target names, I key the mike again: "We should call one of them 'Ishmael,'" I point out to appreciative laughter. The preliminary IDD target gets that name, though later, based on advice from the RAT team, we switch to a different target, Ahab. Well, that's OK: if you're going to be obsessed with a target for a few days, it might as well be a target named for a character who's practically a synonym for obsession.
I finish both sols' IDD sequences and hand them over to Chris. Then I check out the images from yestersol, which include some very cool MIs. The ones we have are slightly out of focus -- not surprising, since we forewent the MB touch in order to preserve the soil -- but we can see that they're going to be awesome when we get the in-focus versions. One looks like a cliff, the others like a miniature city or something.
Just before the CAM I step out to stretch and run into Mark Maimone. He fills me in on Opportunity's progress. They're seeing 99.5% slip: so far they've commanded 48m and have made 27cm of progress. This in itself is more or less what we'd expect, based on the testbed data. But what worries him is that we don't seem to have risen any yet -- it's as if we're just pushing this stuff in front of us all the way down the side of the ripple. He's concerned that we'll end up at the bottom of the ripple and still be buried, in which case we might never get back above the stuff.
On a lighter note, he points out that our expectations have dropped four orders of magnitude. "We used to do 200m sols, now we do 2cm sols," he says. And we're just as happy to get them.
[Next post: sol 497, May 27.]
Courtesy NASA/JPL-Caltech. The view from the end of the drive, with Pequod just in reach.
2010-05-24
Spirit Sol 494
Even though we're not getting the downlink until late -- so we could be starting at a decent hour -- I've had to come in early to support a 10:00 planning meeting. (For some reason, we have this pre-planning meeting at 10:00, even though the uplink process doesn't start until 13:00.) I arrive more than an hour early, because if you show up on Lab past 09:00 all the parking is gone. And I get everything nicely set up, and start some work of my own, and my damn machine reboots.
Turns out they're having some problem with the file servers and they've decided to try to fix it by rebooting all the clients. Without warning.
Well, it's not entirely true that they gave no warning. Apparently, they sent some people email (not me). So if you happened to be one of those people, and you happened to check your email at the right time -- you know, because everybody checks their email every five minutes, just in case the SAs are going to reboot the machines for some incomprehensible reason -- then you had some warning. I guess. The rest of us are supposed to just deal with it, apparently.
Perhaps my attitude is slipping. I need John Wright to come by and grin and say, "This is so cool."
So I get everything back into some semblance of decent shape before the meeting starts. Squyres, despite a badly stuffed-up nose, takes a few minutes to say, "We're shortly going to be facing a big decision on Spirit: how to continue exploring Husband Hill, in light of our recent discoveries here and our improved power situation." He's been mentioning the need for this discussion for a while now, planting the seed. I think this means: "I want to meander up the hillside more slowly than we had been planning to," which is really fine with me. As long as we go to the top.
The pre-planning meeting itself explores a number of options. Are we going to stay here and IDD? Do a long drive back uphill? Do a short drive and explore another part of the outcrop? It's the last choice that wins. We pick a spot at one edge of the outcrop, about 2m from our present position. They'd like to go farther away, but that's where the outcrop ends.
So we'll do one last bit of IDD work on the current chunk of outcrop, then drive on to set ourselves up to continue at another part. I have both sequences roughed out before the SOWG meeting ends, and Chris and I split them: he takes the IDD sequence, I take the drive.
It seems like a long time since I've been RP-1 for a meaningful drive in the Spirit World. It's a short one, but I'm still worried about it. Lots of places where things could go wrong, unknown terrain under our wheels, a long approach we're trying to nail in one sol ... heck, come to think of it, it's just like old times.
Turns out they're having some problem with the file servers and they've decided to try to fix it by rebooting all the clients. Without warning.
Well, it's not entirely true that they gave no warning. Apparently, they sent some people email (not me). So if you happened to be one of those people, and you happened to check your email at the right time -- you know, because everybody checks their email every five minutes, just in case the SAs are going to reboot the machines for some incomprehensible reason -- then you had some warning. I guess. The rest of us are supposed to just deal with it, apparently.
Perhaps my attitude is slipping. I need John Wright to come by and grin and say, "This is so cool."
So I get everything back into some semblance of decent shape before the meeting starts. Squyres, despite a badly stuffed-up nose, takes a few minutes to say, "We're shortly going to be facing a big decision on Spirit: how to continue exploring Husband Hill, in light of our recent discoveries here and our improved power situation." He's been mentioning the need for this discussion for a while now, planting the seed. I think this means: "I want to meander up the hillside more slowly than we had been planning to," which is really fine with me. As long as we go to the top.
The pre-planning meeting itself explores a number of options. Are we going to stay here and IDD? Do a long drive back uphill? Do a short drive and explore another part of the outcrop? It's the last choice that wins. We pick a spot at one edge of the outcrop, about 2m from our present position. They'd like to go farther away, but that's where the outcrop ends.
So we'll do one last bit of IDD work on the current chunk of outcrop, then drive on to set ourselves up to continue at another part. I have both sequences roughed out before the SOWG meeting ends, and Chris and I split them: he takes the IDD sequence, I take the drive.
It seems like a long time since I've been RP-1 for a meaningful drive in the Spirit World. It's a short one, but I'm still worried about it. Lots of places where things could go wrong, unknown terrain under our wheels, a long approach we're trying to nail in one sol ... heck, come to think of it, it's just like old times.
2010-05-18
Spirit Sol 488
"Good, but not perfect." That's Chris's assessment of yestersol's drive. The first guarded arc, meant to take us safely down a smooth slide to our final position, refused to move. This later turns out to be a combination of dangers from nearby hazards -- a couple of rocks beside the top of the slide that are near, or barely above, the hazard-avoidance code's safety threshold -- and an unknown patch on the slide itself, a result of the slide being too smooth for the stereo correlator to determine distance.
So we're about 7m away from our goal. It should be an easy drive thisol, and luckily I prepped the science team in advance, warning them that it might take us three sols to get there. I'm learning.
Today I find out why the science team is so eager to return to Larry's Lookout. From this side, they can see that it's volcanic rock that was altered by water. Not only is water what we came to find, but this alteration is specifically interesting because this rock is tens of meters above the surrounding terrain. That means the water here at Gusev Crater wasn't shallow, and probably wasn't just embedded in a light but long-term sulfuric acid mist (one of the theories I've heard before). This stuff was deep -- tens of meters deep at least. This could be the evidence we were hoping for all along, the evidence of an actual lake here at Gusev.
Tomorrow we should be at the base of the lookout, maybe another sol to climb to the rocks. Then we'll know.
[Next post: sol 494, May 24.]
Courtesy NASA/JPL-Caltech. The view out our front windshield as we planned thisol ....
Courtesy NASA/JPL-Caltech. ... and the corresponding view out the back.
So we're about 7m away from our goal. It should be an easy drive thisol, and luckily I prepped the science team in advance, warning them that it might take us three sols to get there. I'm learning.
Today I find out why the science team is so eager to return to Larry's Lookout. From this side, they can see that it's volcanic rock that was altered by water. Not only is water what we came to find, but this alteration is specifically interesting because this rock is tens of meters above the surrounding terrain. That means the water here at Gusev Crater wasn't shallow, and probably wasn't just embedded in a light but long-term sulfuric acid mist (one of the theories I've heard before). This stuff was deep -- tens of meters deep at least. This could be the evidence we were hoping for all along, the evidence of an actual lake here at Gusev.
Tomorrow we should be at the base of the lookout, maybe another sol to climb to the rocks. Then we'll know.
[Next post: sol 494, May 24.]
Courtesy NASA/JPL-Caltech. The view out our front windshield as we planned thisol ....
Courtesy NASA/JPL-Caltech. ... and the corresponding view out the back.
2010-05-17
Spirit Sol 487
As I'm driving in I see the asshole who got thrown out of this year's JPL Open House for shouting abuse at Mark Adler and Nagin Cox. This happened day before yesterday. Mark and Nagin were standing with the testbed rover outside the 303 cafeteria (where most of the Mars stuff was), volunteering their personal time to answer questions asked by visitors who were in the line to get into the building. I watched this for a while (indeed, late in the day, I did it myself for a while) and most of the visitors were polite, friendly, and supportive. Just nice folks, spending the day there with their kids, learning about space and science.
But not this guy. This jackass was shouting at the top of his lungs, personally insulting Mark and Nagin, calling them liars. Some of the visitors, perhaps worried for their kids, called the guards, and although Nagin pleaded with them not to throw the guy out -- she didn't want anyone to think they weren't free to ask any question they wanted -- I saw them talking to him later on, and shortly after that I heard he had been asked to leave. Now he's standing out by the main road into JPL, where he's put up posters directly and specifically accusing Tim Parker -- by name -- of being a liar, of doctoring images.
I ought to be amused by this. I want to run him over. But if aikido has taught me anything, it's that you don't confront anger with anger, attack with attack. You smile, welcome your attacker's momentum as a gift ... and then spin like a Martian dust devil and throw that motherfucker headfirst into a wall.
If he gets up and comes back, you smile and bow a little and do it again.
See you next year, asshole.
The weekend went well for the rovers. Friday's attempt to drive Opportunity made more progress than expected -- we're still talking about only a couple of centimeters, but we'll take it. And Spirit's IDD work went nominally, too.
So thisol we're finishing up the IDD work, then stowing the IDD and driving the remaining 30m or so to Larry's Lookout. It's a complex sol, so we break up the work: Chris does the drive, I do the IDD stuff. The IDD stuff is actually not too hard, although it does have a twist: the MI images we got back over the weekend suggest that our best-focus position on this surface might not be where we thought, so I work with Craig Leff to come up with a custom strategy.
Chris has the hard part, and he turns out to be the tall pole for the day. I think John Grant regrets that thisol is the sol we chose to drive, because he's got tickets for the Washington Nationals game. When we're done, the game's already in the third inning.
But he makes it.
But not this guy. This jackass was shouting at the top of his lungs, personally insulting Mark and Nagin, calling them liars. Some of the visitors, perhaps worried for their kids, called the guards, and although Nagin pleaded with them not to throw the guy out -- she didn't want anyone to think they weren't free to ask any question they wanted -- I saw them talking to him later on, and shortly after that I heard he had been asked to leave. Now he's standing out by the main road into JPL, where he's put up posters directly and specifically accusing Tim Parker -- by name -- of being a liar, of doctoring images.
I ought to be amused by this. I want to run him over. But if aikido has taught me anything, it's that you don't confront anger with anger, attack with attack. You smile, welcome your attacker's momentum as a gift ... and then spin like a Martian dust devil and throw that motherfucker headfirst into a wall.
If he gets up and comes back, you smile and bow a little and do it again.
See you next year, asshole.
The weekend went well for the rovers. Friday's attempt to drive Opportunity made more progress than expected -- we're still talking about only a couple of centimeters, but we'll take it. And Spirit's IDD work went nominally, too.
So thisol we're finishing up the IDD work, then stowing the IDD and driving the remaining 30m or so to Larry's Lookout. It's a complex sol, so we break up the work: Chris does the drive, I do the IDD stuff. The IDD stuff is actually not too hard, although it does have a twist: the MI images we got back over the weekend suggest that our best-focus position on this surface might not be where we thought, so I work with Craig Leff to come up with a custom strategy.
Chris has the hard part, and he turns out to be the tall pole for the day. I think John Grant regrets that thisol is the sol we chose to drive, because he's got tickets for the Washington Nationals game. When we're done, the game's already in the third inning.
But he makes it.
2010-05-14
Spirit Sol 484
Thisol we're planning a little light IDD work for the weekend -- an MI mosaic, a RAT brush, a RAT-to-APXS tool change. The next sol, we just switch from the APXS to the MB.
Since I'm coming back from the other side of the planet, I have no context at all. The SOWG meeting keeps me hopping, evaluating terrain and imaging requests as well as target reachability. This is the sort of thing that used to drive me nuts back in the nominal mission, but now I have more experience -- and confidence -- so I handle it with aplomb. Maybe even two plombs. Whatever those are.
Later on I find that the target we're using isn't quite as good as I thought earlier. At the end of the first sol, our APXS placement brings the turret right up to the hardstop. The maximum angle for the turret is 3.30 radians, and RSVP shows this placement goes to 3.297. If you allow for the possibility of early contact (a possibility we must always consider), the simulation gives an even more worrisome result: 3.299. Theoretically we should be fine, but that's not exactly a whole lot of margin. So I end up tweaking the target normal slightly, just for the APXS placement.
It's Jascha's last sol before he goes off to grad school, and it's Heather Arneson's last sol as a PANCAM PUL; in two weeks, she'll also be gone. I never knew Jascha well, but I knew Heather very well. She was not only damn good at her job, she also had a fantastic attitude. I missed her when she left JPL to go back for Ithaca -- for months she's been just a voice on the phone -- and I'll miss her when she leaves us for good. As it happens, Heather (and Miles, her boyfriend and also a PANCAM team member) is heading to my own alma mater, the UIUC. "It's a great school, if you can handle the weather," I tell her. Then again, she's used to Ithaca, NY. Heck, Illinois weather might be better.
Shoot, Mars might be better.
[Next post: sol 487, May 17.]
Since I'm coming back from the other side of the planet, I have no context at all. The SOWG meeting keeps me hopping, evaluating terrain and imaging requests as well as target reachability. This is the sort of thing that used to drive me nuts back in the nominal mission, but now I have more experience -- and confidence -- so I handle it with aplomb. Maybe even two plombs. Whatever those are.
Later on I find that the target we're using isn't quite as good as I thought earlier. At the end of the first sol, our APXS placement brings the turret right up to the hardstop. The maximum angle for the turret is 3.30 radians, and RSVP shows this placement goes to 3.297. If you allow for the possibility of early contact (a possibility we must always consider), the simulation gives an even more worrisome result: 3.299. Theoretically we should be fine, but that's not exactly a whole lot of margin. So I end up tweaking the target normal slightly, just for the APXS placement.
It's Jascha's last sol before he goes off to grad school, and it's Heather Arneson's last sol as a PANCAM PUL; in two weeks, she'll also be gone. I never knew Jascha well, but I knew Heather very well. She was not only damn good at her job, she also had a fantastic attitude. I missed her when she left JPL to go back for Ithaca -- for months she's been just a voice on the phone -- and I'll miss her when she leaves us for good. As it happens, Heather (and Miles, her boyfriend and also a PANCAM team member) is heading to my own alma mater, the UIUC. "It's a great school, if you can handle the weather," I tell her. Then again, she's used to Ithaca, NY. Heck, Illinois weather might be better.
Shoot, Mars might be better.
[Next post: sol 487, May 17.]
2010-05-13
Opportunity Sol 463 (Spirit Sol 483)
I'm back on Spirit as of today, but yesterday they started a two-sol MB integration. So I basically have nothing to do. (But I do have one nice surprise at the SOWG meeting: I overhear Brenda say, "Even Squyres admitted the other day that Spirit is now his favorite.")
So I do some other work and then go hang out with the Opportunity drivers. The good news is, the wheels steered, exactly the way we wanted. Jeff and Frank are mucking around with the candidate sequence Khaled and I wrote yesterday (they almost seem apologetic about it, but I tell them, "Mess with it to your heart's content"). And just after I poke my head in, there's a meeting about the Opportunity mobility plan.
I tag along for the meeting. Khaled's there, along with Jeng and Mark Maimone. When Jeff, Frank, and I show up, I count heads. Counting Mark as one, that makes six rover drivers -- one for each wheel. Not that we'd ever overdo things around here.
The meeting consists of getting everybody (RPs, managers, scientists, mechanical people, press people, and more) together in the SOWG room to tell them that the wheels steered and that we're going to try to drive.
Well, all right, there's a little more to it than that. The real point of the meeting is to go over the details. Mark is the main one doing the analysis, and is, as usual, amazingly thorough. Plot after plot, image after image, he methodically lays out the entire sequence of events. Which can be summarized this way: we turned the wheels, and sank a tiny amount (as expected) when we did that. And we're ready to try driving.
So I do some other work and then go hang out with the Opportunity drivers. The good news is, the wheels steered, exactly the way we wanted. Jeff and Frank are mucking around with the candidate sequence Khaled and I wrote yesterday (they almost seem apologetic about it, but I tell them, "Mess with it to your heart's content"). And just after I poke my head in, there's a meeting about the Opportunity mobility plan.
I tag along for the meeting. Khaled's there, along with Jeng and Mark Maimone. When Jeff, Frank, and I show up, I count heads. Counting Mark as one, that makes six rover drivers -- one for each wheel. Not that we'd ever overdo things around here.
The meeting consists of getting everybody (RPs, managers, scientists, mechanical people, press people, and more) together in the SOWG room to tell them that the wheels steered and that we're going to try to drive.
Well, all right, there's a little more to it than that. The real point of the meeting is to go over the details. Mark is the main one doing the analysis, and is, as usual, amazingly thorough. Plot after plot, image after image, he methodically lays out the entire sequence of events. Which can be summarized this way: we turned the wheels, and sank a tiny amount (as expected) when we did that. And we're ready to try driving.
2010-05-12
Opportunity Sol 462 (Spirit Sol 482)
It's sol 462 here in the Land of Opportunity -- or, rather, we're planning sol 462. But we still haven't uplinked sol 461, because we're in that weird phase.
It's a restricted sol, which means we RPs have no sequences to build. The SOWG meeting finishes in record time (literally).
We have no idea whether yestersol's wheel-steering sequence worked, but Khaled and I plan a candidate drive sequence anyway. We know the general outlines of what Jeff wants to try, and we build that: a 2-meter arc back the way we came, with a heading change to the left, broken up into 10 steps of 20cm each. And lots and lots and lots of pictures.
Will we send it? We'll have to wait until we see what happened when we tried to steer.
It's a restricted sol, which means we RPs have no sequences to build. The SOWG meeting finishes in record time (literally).
We have no idea whether yestersol's wheel-steering sequence worked, but Khaled and I plan a candidate drive sequence anyway. We know the general outlines of what Jeff wants to try, and we build that: a 2-meter arc back the way we came, with a heading change to the left, broken up into 10 steps of 20cm each. And lots and lots and lots of pictures.
Will we send it? We'll have to wait until we see what happened when we tried to steer.
2010-05-11
Opportunity Sol 461 (Spirit Sol 481)
Frank Hartman, Jeng Yen, and Khaled Ali are already at the SOWG meeting when I show up, making me the fourth RP in the room. All this for a drive sequence that just steers the wheels to a different position. "We're multiplying," Frank says. "We can each take one milliradian of the turn."
Well, with so many of us around, and the bulk of the work already done yesterday, I suppose I won't have much to do. I do get involved in a discussion about some drive-supporting PANCAM imaging. Blue-stereo images usually give us the sharpest picture, but Elaina McCartney points out that the images will be taken at a particularly warm time of day, and consequently the blue-stereo images will have speckling that might interfere with the analysis we're trying to do. So we switch to the next-best thing, red-stereo. But she also generously volunteers to take a blue-stereo pair as well, just so we can see what the images would look like, in case they might prove useful after all. Despite the large data volume required for this, the SOWG chair, Ed Guinness, shrugs acceptance. "Damn," I say, "it's almost like we're a team here or something."
My easy, laid-back day turns out to be a lot busier than I anticipated. Frank and Jeng go down to the testbed to start testing out the next sequence, and Khaled and I have more work to do on thisol's sequence than we'd thought. Nothing major, just lots of picky little things. Working out the final image pointing. Mucking with the visodom image parameters so we'll get more data. Going over the timing with the TULs. That sort of thing.
Indeed, when lunchtime approaches, we're both too busy to leave. "We need a gofer," I say, and Khaled agrees wholeheartedly, or rather wholestomachedly.
But we're ready in time for all the usual meetings. (I even show an animation at the APAM: the rover sits still for five seconds, then turns the wheels. Woo-hoo. I attach the animation to the uplink report.)
When it's all said and done, not only all four RPs who were there at the SOWG but also Jeff Biesiadecki and Mark Maimone have reviewed it, along with many others, such as Rick Welch. If it takes all this just to steer the wheels now, I don't want to see what happens when we try to drive. It's a far cry from the days when sequencing Opportunity meant: well, we expanded the über-macro, there's your 200m drive -- see ya!
Well, with so many of us around, and the bulk of the work already done yesterday, I suppose I won't have much to do. I do get involved in a discussion about some drive-supporting PANCAM imaging. Blue-stereo images usually give us the sharpest picture, but Elaina McCartney points out that the images will be taken at a particularly warm time of day, and consequently the blue-stereo images will have speckling that might interfere with the analysis we're trying to do. So we switch to the next-best thing, red-stereo. But she also generously volunteers to take a blue-stereo pair as well, just so we can see what the images would look like, in case they might prove useful after all. Despite the large data volume required for this, the SOWG chair, Ed Guinness, shrugs acceptance. "Damn," I say, "it's almost like we're a team here or something."
My easy, laid-back day turns out to be a lot busier than I anticipated. Frank and Jeng go down to the testbed to start testing out the next sequence, and Khaled and I have more work to do on thisol's sequence than we'd thought. Nothing major, just lots of picky little things. Working out the final image pointing. Mucking with the visodom image parameters so we'll get more data. Going over the timing with the TULs. That sort of thing.
Indeed, when lunchtime approaches, we're both too busy to leave. "We need a gofer," I say, and Khaled agrees wholeheartedly, or rather wholestomachedly.
But we're ready in time for all the usual meetings. (I even show an animation at the APAM: the rover sits still for five seconds, then turns the wheels. Woo-hoo. I attach the animation to the uplink report.)
When it's all said and done, not only all four RPs who were there at the SOWG but also Jeff Biesiadecki and Mark Maimone have reviewed it, along with many others, such as Rick Welch. If it takes all this just to steer the wheels now, I don't want to see what happens when we try to drive. It's a far cry from the days when sequencing Opportunity meant: well, we expanded the über-macro, there's your 200m drive -- see ya!
2010-05-10
Opportunity Sol 460 (Spirit Sol 480)
Yikes. Cooper had an emergency appendectomy this weekend. He's consequently unable to come to work, so I volunteer to fill in for him. This puts me back on Opportunity for the first time in quite a while -- today, and for the next couple of days.
The Opportunity team has just about decided to drive out of Purgatory. They haven't even tried moving yet, but testbed testing has been encouraging. They've been able to simulate most of the important conditions of the real vehicle, and they keep managing to drive it out.
Indeed, we're back in the testbed again today. Jeng and I whip up a candidate sequence, whose only mobility command is to steer the wheels in preparation for the first stage of the drive. This single command is surrounded (fore and aft) by a zillion imaging commands. We're taking this very, very slowly.
Then we walk down to the testbed, where we meet up with Frank and Mark. We won't be driving the testbed rover in the special soil they cooked up to simulate Opportunity's current site, due to medical concerns: the fine particulates are dangerous to breathe for long periods, and we haven't been formally trained on the respirators. (I don't honestly see what's to know. It's a face mask. You strap it on. But their intent is to keep us healthy, and I can't find too much fault with that. They're insane, but they mean well.[1])
As usual, it just takes forever to get anything done. A couple of hours go by, and we haven't run the sequence yet. Both Frank's and my ESD certifications have lapsed, so we can't man the kill switch anyway, even once the rover gets going. We mostly stand there chit-chatting -- about the rovers, about LISP, about camera phones, about whatever. An employee I've never met brings a couple of guests by, and I go over and talk with them for a while about Opportunity's current status and what we're working on. "That's the second time I've seen you do that," Frank says after they depart. "You're like a rover ambassador."[2]
Then, just before things really get going, I have to leave. I'm working the Open House this weekend, and they have an orientation meeting this afternoon. Which is mostly useless, but one of the speakers makes a good point: "Not everyone has the privilege of working somewhere where everyone wants to go there and see what they do." True, and that's why I'm doing the Open House: for the sake of all those poor slobs who want to be me.
But seriously, I am impressed with the scale of the effort. As another speaker puts it, "This place was not set up to be Disneyland." But we turn it into Disneyland for a couple of days a year, and that's pretty darn cool, really.[3] I'm only too happy to be a part of it again. (Also, she mentions offhandedly that they used to have an open house on the last Sunday of every month! I can't imagine it was anything like the scale of the present undertaking, but even so, that must have been a very different time.)
I make it back to the testbed just after the sequence finishes running, which pretty much officially means I wasted my entire day. We then have a long conference call with the rest of the uplink team, to set up for tomorrow. They'll be doing additional testing on this sequence tomorrow in parallel with the uplink process, but it's clear that at least some people think we're jumping through a ridiculous number of hoops: as Jeff Biesiadecki puts it, "The rover would have to catch on fire for us not to send this sequence."[4]
So I guess we're going ahead with that, then.
I go back to my office and review the sequence once again, and I do catch a couple of minor errors -- one of which would have prevented Opportunity from moving at all. So I guess it wasn't a complete waste after all.
Footnotes:
[1] I eventually had the face-mask training, as part of the Spirit extrication effort. Turns out it's a face mask, and you strap it on.
OK, so there's a little more to it than that. But not much.
[2] To this day, that was the nicest way anyone's ever called me an attention whore.
[3] This is probably why I started calling it "Deep Space Disneyland."
[4] What sequence do we send if that happens?!
The Opportunity team has just about decided to drive out of Purgatory. They haven't even tried moving yet, but testbed testing has been encouraging. They've been able to simulate most of the important conditions of the real vehicle, and they keep managing to drive it out.
Indeed, we're back in the testbed again today. Jeng and I whip up a candidate sequence, whose only mobility command is to steer the wheels in preparation for the first stage of the drive. This single command is surrounded (fore and aft) by a zillion imaging commands. We're taking this very, very slowly.
Then we walk down to the testbed, where we meet up with Frank and Mark. We won't be driving the testbed rover in the special soil they cooked up to simulate Opportunity's current site, due to medical concerns: the fine particulates are dangerous to breathe for long periods, and we haven't been formally trained on the respirators. (I don't honestly see what's to know. It's a face mask. You strap it on. But their intent is to keep us healthy, and I can't find too much fault with that. They're insane, but they mean well.[1])
As usual, it just takes forever to get anything done. A couple of hours go by, and we haven't run the sequence yet. Both Frank's and my ESD certifications have lapsed, so we can't man the kill switch anyway, even once the rover gets going. We mostly stand there chit-chatting -- about the rovers, about LISP, about camera phones, about whatever. An employee I've never met brings a couple of guests by, and I go over and talk with them for a while about Opportunity's current status and what we're working on. "That's the second time I've seen you do that," Frank says after they depart. "You're like a rover ambassador."[2]
Then, just before things really get going, I have to leave. I'm working the Open House this weekend, and they have an orientation meeting this afternoon. Which is mostly useless, but one of the speakers makes a good point: "Not everyone has the privilege of working somewhere where everyone wants to go there and see what they do." True, and that's why I'm doing the Open House: for the sake of all those poor slobs who want to be me.
But seriously, I am impressed with the scale of the effort. As another speaker puts it, "This place was not set up to be Disneyland." But we turn it into Disneyland for a couple of days a year, and that's pretty darn cool, really.[3] I'm only too happy to be a part of it again. (Also, she mentions offhandedly that they used to have an open house on the last Sunday of every month! I can't imagine it was anything like the scale of the present undertaking, but even so, that must have been a very different time.)
I make it back to the testbed just after the sequence finishes running, which pretty much officially means I wasted my entire day. We then have a long conference call with the rest of the uplink team, to set up for tomorrow. They'll be doing additional testing on this sequence tomorrow in parallel with the uplink process, but it's clear that at least some people think we're jumping through a ridiculous number of hoops: as Jeff Biesiadecki puts it, "The rover would have to catch on fire for us not to send this sequence."[4]
So I guess we're going ahead with that, then.
I go back to my office and review the sequence once again, and I do catch a couple of minor errors -- one of which would have prevented Opportunity from moving at all. So I guess it wasn't a complete waste after all.
Footnotes:
[1] I eventually had the face-mask training, as part of the Spirit extrication effort. Turns out it's a face mask, and you strap it on.
OK, so there's a little more to it than that. But not much.
[2] To this day, that was the nicest way anyone's ever called me an attention whore.
[3] This is probably why I started calling it "Deep Space Disneyland."
[4] What sequence do we send if that happens?!
2010-05-06
Spirit Sol 477
Yestersol's drive was too complex. I was thinking about it last night, and my hopes sank lower and lower. We should have broken it into multiple sols. We shouldn't have tried to thread the needle at the end. I should have looked more carefully for obstacles.
Of course, it went splendidly. (If I learned anything from experience, I'd have been expecting this. But I never do.) Dan Moyers tells me the good news as I walk into the SOWG meeting. "Excellent!" I reply. "I was all ready to blame Chris if it went wrong."
As has become our Friday custom, we're planning the rover's activities for the whole weekend. There's lots of IDD work, and lots of remote sensing. (In keeping with the nautical theme of Jib Sheet, most of the remote-sensing targets get names such as "gunwale," "keel," etc. They're taking some pictures of the tracks as well, and Charles Budney suggests the name "wake" for that target.) We have a limited time -- not only are we supposed to stick to a 10-hour day anyway, but uplink's at 19:00, which pretty much enforces that limit thisol. As a result of this tight schedule, Dan rejects a request from the atmospheres team -- it'll just be too hard to fit it in in the time we've got, given the complexity of the overall plan.
As you can imagine, the atmospheres team isn't too happy about that, and they push back hard. But Dan sticks to his guns, and Emily backs him up. "The resource we need to optimize here is team planning time," she says. One of the scientists turns to me off mike and says, "So the engineering team is no longer competent to --" but he's interrupted, and he never finishes the thought. Just as well.
In the room behind us, the Opportunity team is having its own debate, joined by some higher management. Opportunity's been stuck on a dune for the last few weeks, buried up to the tops of the wheels -- almost up to the bottom of the WEB -- and management's been paranoid about allowing us to drive it out. I notice that Jay Torres was sitting by a window that afforded a view into that room, so I ask him if he saw any of the presentation. He shakes his head no, as Chris chimes in, "The content is, we can be a lander or we can drive backward a couple of meters."
"Which we knew three weeks ago," I say.
"Yeah," he replies, "but now our ass is covered."
Yeah, that probably sums it up.
Back downstairs, Emily asks our new TDL what he thought of the acrimony in the SOWG meeting. "I can see why they keep the scientists and engineers in physically separate locations," he says diplomatically.
"Well, it used to be better," Emily says. "But the vehicles have gotten harder to operate. We used to have no power, so we had no concerns about thermal issues or planning time -- there was no power to do anything. But now we're power-rich again, and we'd forgotten there were other constraints."
This is true, but I think it's incomplete. Another part of the problem is that the scientists just aren't here any more. They're voices on a phone, and that makes it easier for both sides to degenerate into an us-versus-them mentality. Worse, they simply don't get to see the whole process any more -- most of them sign off when they're done with their part of the planning, but we keep working for hours. If they can't see what effect their decisions have on the sequencing team, how can they hope to make wise decisions about what can and can't get done? I point this out to Emily, and semi-jokingly suggest a solution: "We should get Squyres to require all of the science team to stay online until we're done. You want us to do some observation that's going to take us two extra hours to fit into the plan? Fine, but you stay here until we leave."
I think Emily likes the idea, but I doubt she'll pitch it to Steve. I might. If the problem gets worse, I might have to.
But at least the rover planners have an easy sol. It's just some light IDD work. Chris is done with the sequences maybe half an hour after the SOWG meeting ends. We hand off and I make some minor changes, but nothing particularly urgent.
Since we're so far ahead of the game, I consider going ahead with the time-consuming process of building an MPEG movie of the planned IDD work for the weekend, which I customarily attach to the sol's uplink report. But when we're this far ahead, something always happens to change the plan, and today's no exception. When we show the animation at the activity plan approval meeting, we find out that the science team changed its mind about which IDD targets they wanted us to use, and somehow the word didn't make it to us. No big deal; that's why we do that. I have it fixed within 15 minutes or so.
But I was right not to go ahead and build the movie early. Huh. Seems like sometimes I do learn from experience, after all.
[Next post: sol 480 (Opportunity sol 460), May 10.]
Of course, it went splendidly. (If I learned anything from experience, I'd have been expecting this. But I never do.) Dan Moyers tells me the good news as I walk into the SOWG meeting. "Excellent!" I reply. "I was all ready to blame Chris if it went wrong."
As has become our Friday custom, we're planning the rover's activities for the whole weekend. There's lots of IDD work, and lots of remote sensing. (In keeping with the nautical theme of Jib Sheet, most of the remote-sensing targets get names such as "gunwale," "keel," etc. They're taking some pictures of the tracks as well, and Charles Budney suggests the name "wake" for that target.) We have a limited time -- not only are we supposed to stick to a 10-hour day anyway, but uplink's at 19:00, which pretty much enforces that limit thisol. As a result of this tight schedule, Dan rejects a request from the atmospheres team -- it'll just be too hard to fit it in in the time we've got, given the complexity of the overall plan.
As you can imagine, the atmospheres team isn't too happy about that, and they push back hard. But Dan sticks to his guns, and Emily backs him up. "The resource we need to optimize here is team planning time," she says. One of the scientists turns to me off mike and says, "So the engineering team is no longer competent to --" but he's interrupted, and he never finishes the thought. Just as well.
In the room behind us, the Opportunity team is having its own debate, joined by some higher management. Opportunity's been stuck on a dune for the last few weeks, buried up to the tops of the wheels -- almost up to the bottom of the WEB -- and management's been paranoid about allowing us to drive it out. I notice that Jay Torres was sitting by a window that afforded a view into that room, so I ask him if he saw any of the presentation. He shakes his head no, as Chris chimes in, "The content is, we can be a lander or we can drive backward a couple of meters."
"Which we knew three weeks ago," I say.
"Yeah," he replies, "but now our ass is covered."
Yeah, that probably sums it up.
Back downstairs, Emily asks our new TDL what he thought of the acrimony in the SOWG meeting. "I can see why they keep the scientists and engineers in physically separate locations," he says diplomatically.
"Well, it used to be better," Emily says. "But the vehicles have gotten harder to operate. We used to have no power, so we had no concerns about thermal issues or planning time -- there was no power to do anything. But now we're power-rich again, and we'd forgotten there were other constraints."
This is true, but I think it's incomplete. Another part of the problem is that the scientists just aren't here any more. They're voices on a phone, and that makes it easier for both sides to degenerate into an us-versus-them mentality. Worse, they simply don't get to see the whole process any more -- most of them sign off when they're done with their part of the planning, but we keep working for hours. If they can't see what effect their decisions have on the sequencing team, how can they hope to make wise decisions about what can and can't get done? I point this out to Emily, and semi-jokingly suggest a solution: "We should get Squyres to require all of the science team to stay online until we're done. You want us to do some observation that's going to take us two extra hours to fit into the plan? Fine, but you stay here until we leave."
I think Emily likes the idea, but I doubt she'll pitch it to Steve. I might. If the problem gets worse, I might have to.
But at least the rover planners have an easy sol. It's just some light IDD work. Chris is done with the sequences maybe half an hour after the SOWG meeting ends. We hand off and I make some minor changes, but nothing particularly urgent.
Since we're so far ahead of the game, I consider going ahead with the time-consuming process of building an MPEG movie of the planned IDD work for the weekend, which I customarily attach to the sol's uplink report. But when we're this far ahead, something always happens to change the plan, and today's no exception. When we show the animation at the activity plan approval meeting, we find out that the science team changed its mind about which IDD targets they wanted us to use, and somehow the word didn't make it to us. No big deal; that's why we do that. I have it fixed within 15 minutes or so.
But I was right not to go ahead and build the movie early. Huh. Seems like sometimes I do learn from experience, after all.
[Next post: sol 480 (Opportunity sol 460), May 10.]
2010-05-05
Spirit Sol 476
As we work our way back down the ridge to Larry's Lookout, our next stop is Jib Sheet. Thisol we're sequencing a drive to take us most of the way there; wherever we end up, we'll IDD over the weekend, and then complete the Jib Sheet approach early next week.
It's a complex drive. To get there, we'll need to drive in a big "U." It's a total distance of about 25m, a much longer drive than we're normally able to accomplish in this challenging terrain. In our favor, the terrain does flatten out considerably as we go, which will make the vehicle's performance more predictable. But the length of the drive and the nature of the overall terrain will make this a long and difficult plan.
We're not sure for a while exactly where we want the rover to end up, or how it will get there, or even whether we'll drive forward or backward. Dan Moyers sees this and decides to have a little fun with TUL-in-training Rich. "We need the final heading so we can do comm predicts," he says. "Ask the rover planners what our final heading will be."
But Rich is on the ball. "Dan wants to know our final heading."
"Sixty," Chris says without turning around. "Plus or minus 180."
It's mostly Chris's show today. I do the stow and work out the best plan for the first couple of meters of the drive, just enough to get us on the road. Chris does most of the work of planning out the route, though I throw in my opinion now and then. For instance, he's hoping to take a shortcut between two larger rocks, letting autonav find a safe path. I do some analysis to work out that the rover won't be able to find a safe path without loosening the autonav settings, and neither one of us wants to do that. So we take the long path.
Our accidental IDD animation keeps turning up new surprises. I composed and posted the full two-sol version of the animation the other day, and Brenda Franklin's been looking over it carefully. Her scrutiny was rewarded with a neat surprise: in the second-to-last frame on the second sol, a huge gust of wind -- maybe a dust devil -- kicks up a huge cloud of dust on a far hill. "Definitely lemonade with whiskey," she says.
(I've had trouble with animated GIFs and Blogger. You should be able to see the animation at twitpic.)
Courtesy NASA/JPL-Caltech. The dust devil is in the upper-left corner, near the horizon. Subtle, but it's there.
It's a complex drive. To get there, we'll need to drive in a big "U." It's a total distance of about 25m, a much longer drive than we're normally able to accomplish in this challenging terrain. In our favor, the terrain does flatten out considerably as we go, which will make the vehicle's performance more predictable. But the length of the drive and the nature of the overall terrain will make this a long and difficult plan.
We're not sure for a while exactly where we want the rover to end up, or how it will get there, or even whether we'll drive forward or backward. Dan Moyers sees this and decides to have a little fun with TUL-in-training Rich. "We need the final heading so we can do comm predicts," he says. "Ask the rover planners what our final heading will be."
But Rich is on the ball. "Dan wants to know our final heading."
"Sixty," Chris says without turning around. "Plus or minus 180."
It's mostly Chris's show today. I do the stow and work out the best plan for the first couple of meters of the drive, just enough to get us on the road. Chris does most of the work of planning out the route, though I throw in my opinion now and then. For instance, he's hoping to take a shortcut between two larger rocks, letting autonav find a safe path. I do some analysis to work out that the rover won't be able to find a safe path without loosening the autonav settings, and neither one of us wants to do that. So we take the long path.
Our accidental IDD animation keeps turning up new surprises. I composed and posted the full two-sol version of the animation the other day, and Brenda Franklin's been looking over it carefully. Her scrutiny was rewarded with a neat surprise: in the second-to-last frame on the second sol, a huge gust of wind -- maybe a dust devil -- kicks up a huge cloud of dust on a far hill. "Definitely lemonade with whiskey," she says.
(I've had trouble with animated GIFs and Blogger. You should be able to see the animation at twitpic.)
Courtesy NASA/JPL-Caltech. The dust devil is in the upper-left corner, near the horizon. Subtle, but it's there.
Subscribe to:
Posts (Atom)
Posts
