Thursday, October 20, 2011

Road Music

Wait! Opportunity! Before you drive North, you'll need some road music.

(Vague reference to TAP/SIE album of the day bSols 2751-2: The Benefit Concert, Vol. 3 — Warren Haynes, et al)

A lesson in the Deep Space Network is forthcoming. Bring your maths. All your maths.

Tuesday, October 18, 2011

Yeah, well… your face.

The round table in the MER-B sequencing room is, well, round. It conveys a sense of control of both the room and the rover. With this comes an inevitable feeling that we — only us JPL'ers, only us in the room — are in control of everything, that we know everything. Of course, we don't. Far from it. If we did, we wouldn't need our SOWG chair (who is ever so kind as to point out that we're here to do science, not backflips), our documentarian (who is ever so kind as to remind us that we forgot to add the cal target image and that we didn't give the target a damned name), our instrument PULs (who are ever so kind to actually build those beautiful imaging sequences), and our coffee mugs (who, contrary popular opinion, can write a pretty mean Perl script).

This realization hits us in the room pretty often. Well, at least me. The rest of them can speak for themselves. The fun (sic) wakeup calls are the ones coming from "over the line" — over the phone. Since the MER project moved to Earth time and the majority of the science team moved back home, far away from JPL to places like New York and St. Louis, we've had to operate with half the team being local to JPL and half the team being remote. Setting up secure computer and phone lines, that's all trivial stuff. What the project really lost, what was really a challenge was having the whole team in the room. We got some of that back with the Polycom, but really only during the SOWG meeting in the mornings. Now, we've lost even that capability. We're stuck with people — our SOWG chair, usually, as well as our documentarian and most of our instrument PULs — being only voices to us.

I'm notorious, if only to myself, for saying stupid things in front of people who know better. My filter on asking questions and puking words went out the door a long time ago, so it's only statistically probable that I'll say something profoundly wrong. I can trace it back to an aircraft dynamics and control course I took in school where suddenly I lost faith in everything because, let's face it, the Dutch Roll response makes no sense, and therefore nothing else should. 

Figure 1: Confused.jpg

I use this picture for two reasons. Well, three. First, I like analogies. They work. They fit things together. I like neat things. Second, it shows that, yeah, that guy up there is showing us hotshot undergrads how life really works (z = e^[st], is how), and we really don't know what we're talking about even when we think we've got the room under control. This is how I feel when I think I know how, say, the Pancam takes and compresses its images, but then the Pancam PUL comes and says, "Well, z = e^[st], so, you're wrong, but it's cool, you just learned something." Third… can you see his face? No, you can't. I promise. This faceless person has an end goal with an end product. In the picture, that end product happens to be a lesson on the principles of linearized aircraft dynamics.* We, the users of the information, are not privy to the data stream from which this information springs. All we see are abstractions of this data stream — letters and numbers. This guy turns his back to the room and just starts going. Like it's his job. (Ahem.)

For us in the room at JPL, we have ourselves a set of voices representing remote participants who are working and massaging their respecting parts of the current tactical plan for Opportunity. They all have a goal with an associated end product. But, to us, they are effectively just voices in the ether. The phone glows green to let us know there are folks — you know, real, actual people — on the line, but it's not really… enough. They are faceless contributors to an extremely complex set of 0's and 1's. Their backs are to us. They're just facing the chalkboard, and all we get are abstractions of their thought process — sequence IDs and strategic plans and rover attitude translations.**

Years back, probably shortly after the transition to remote operations for the non-JPL'ers, the JPL team got pictures of all of the SOWG Chairs and taped them plastic spoons. To this day, when a given SOWG chair is on duty for the day, we get their face and place it in cup with an LA Kings jersey on it. There they are: our SOWG chair, LA Kings fan and all. (Hey, at least it's not the Red Wings.)

The pictures-on-a-jersey method gets the trick done to some extent. But it's not enough; I don't suppose that Ray Arvidson is actually smiling all the time, and yet his picture depicts this. Again, it's just an abstraction. We can't see his body language. When someone is right next to me, in the room, I can see the way they lean into the microphone, or their body language when someone else on the line says something intriguing, or confusing, or wrong, or right. Chair fidgeting, though disconcerting to those around you, can be very telling. This is important, this learning to read people's body language thing, and it makes it so much harder to really get someone's stake in a given moment or event when you can't see how they're reacting.

Well, it is what it is. The remote participation is pretty cool, actually: We've got people in three different time zones coming together to make a piece of metal on Mars do what we tell it. Ridiculous. Sometimes it helps to imagine these voices as something representing their stake in the tactical plan, like a piece of hardware, plopped on top of a rudimentary stick figure:

Figure 2: Crude and ill-conceived representation of MER Remote Participant; not to scale. Also totally inaccurate.

So, we've got this person on the phone, a stick figure with a camera. And I say something silly. Despite all the abstractions and walls between our lines of communication, it is very easy to show me that I said something silly. When someone tells me that I'm wrong, and it's clear to me that I am, all I want to say is:

Yeah, well… your face.

Which, you know, isn't really a response. Turns out. Since, you know, I don't know what their face looks like. Turns out. And even if I did, then I'd be, you know, still wrong. Which is cool, because I don't get stepped on by anyone but myself. And plus, I just learned something. Even cooler.

So, sorry Pancam Nancy and Ecam Billy Bob. I just see a camera mast on a stick figure. Maybe we should get jersey'd cups for you guys, too?


By the way, no albums of the day for now, since I haven't been on tactical for a week and a half… but tomorrow, I shall rain the heaven of music upon thy ground.

*F-16 on brick times brick does not equal F-16. Turns out.
*The analogy breaks down, of course, when you remind yourself that the teacher occasionally turns around, and we get to see his face.

Tuesday, October 11, 2011

A Day In the Life, bSols 2742-I Don't Even Know Anymore: On the Road Again

Hey. Everybody.


We're driving again!

Or, for the label and topography inclined:

As some have correctly guessed (or… knew…), Opportunity has been on the move. In fact, we've been moving so quickly that I had lost a lot of the context for why we were suddenly moving. I wouldn't call it a sense of urgency, just prudence. 

There are two very simple things that we, as the MER tactical team, like to see:

1) A rover that is alive
2) A rover than is doing science

#1 always takes precedence. Always. 

#2, of course, isn't just a nice-to-have, it's a must-have. ("Design is based on requirements.")

How do we reconcile these two things? Very, very carefully. The current context is that the "countdown to winter" clock was unofficially started a few weeks ago, meaning we had to boogie on out of the parking spot at Chester Lake in a campaign that is primarily about finding north-facing slopes. We don't need to be there for a few months, as winter solstice is on March 31. Here, I'll prove it. With math. Well, with SPICE*. So I'm cheating. But not really.

(Rule #319: SPICE is everybody's friend!)**

Although it doesn't look like it, the southern hemisphere of Mars — Oppy's hemisphere — has its winter when it is farthest from the sun. (And, of course, its summer happens to occur when Mars is closest to the sun. The same thing occurs on Earth, incidentally. Summers in the southern hemisphere are noticeably harsher because of just that little bit that the Earth is closer to the sun during that time. Recall that season is only defined by tilt relative to the Sun, not distance to the Sun!) Here, take this plot of Sun-Mars range over time:

(Rule #320: SPICE is still everybody's friend!)

Ok, so, right, fancy plots. So what? 

The point at which we need to find ourselves some nice parking spots with north-facing slopes isn't just at the tip of the plot (March 31). It's a time period a little before then through a little after then. Where, then, are our options? We could go north, and we could go south. Here — context image! Take it! You'll need it!

To the north is more Cape York, and some great candidates for both science and north-facing slopes. To the south, our long-term targets of Sutherland Point, Knobby's Head, and Solander Point. Sutherland and Knobby are reachable by the time we need to be concerned, but from the looks of it, they don't have any lily pads worth our time (or our rover). As I've pointed out previously, there's still science to be done at Cape York, where the phyllosilicates are waiting to take an atomic beating from our Mossbauer. So, the game we're playing is to find a spot — or series of spots — that first and foremost protects our rover by tilting her for solar juice and that has exposures for science collection. MB'ing is not power intensive, so conducting the campaign of "mapping" Cape York now is the best choice. Given the drive path and its cadence with the sol numbers, we can expect limited science collection with the IDD until we find a spot to catch our breath. 

What we get, though, are image sets that remind us of the long trek to Endeavour: Pancams and Navcams galore. Our Pancam and Ecam PULs are pulling the trigger on new sequences every day. Although Cape York is much more bland 

Stolen from fredk and Jan van Driel @ UMSF!

We're also getting some interesting drive paths, as you can see in the "V" drive at the end — we saw that small, ~15-meter diameter crater and decided to do a drive-by shot to characterize its southern bits, which face north, and then head back Northeast. 

Expect a busy bit of driving and exploration, despite the move into Restricted planning starting this Friday — a drive every 3 sols or so will be the norm, unless we find aliens. In which case… meh. 


*Follow that link and you'll be inundated with kernels. How do you use SPICE? Well… that will take a while. Gotta know orbital mechanics and gotta know programming. In any case, the lesson here is that you can learn an extraordinarily huge amount of stuff about the planets and all of JPL's spacecraft with SPICE, and it's all free and open to the public. Want the code? I did it in MATLAB (an old student version, too!), since I'm awesome. 
**Bonus points if you can tell me why everything is tilted into the Z-plane. I did the plots in the J2000 frame, for the record…

Literally unbelievable

The long-awaited Endeavour Movie has been released:

For perspective, I joined the project near sol 2320 — right when the view gets good.


Update post coming soon! I found 30 minutes to spare!

Thursday, October 6, 2011

Oh, you're looking at *me*

Every month or so, the IST Team Chief sends out an email with a link to giant Excel spreadsheet. This spreadsheet contains the tactical shift schedule for everybody on the team, from MMs to TAP/SIEs to RPs to team ninjas. With the email comes a few comments. This month's set of comments included an interesting bit:

Matt Lenda will begin TUL training

Time to strap my learning hat on. It's almost as if I'm starting all over again.


Each flight project at JPL has a very specific way of "doing business." Everyone has their own flavors, their own takes on things, their own politics, their own tools and processes. MER does not stray from this trend — it is the same in that it is different from every other project. One striking feature of this difference is that each project will have a peculiar spot where they "bury the bodies." 

So to speak.

At JPL, we have a particular process architecture for going from ideas ("I wanna do science stuff") to activities ("I wanna use this thing to do my science") to sequences ("I wanna use these commands and directives to tell the thing to do my science"). This architecture is roughly the same for all flight projects that JPL manages. "Burying the bodies" means that everybody puts their own spin on this process. The same idiosyncrasies rear their ugly (or very, very pretty) head in slightly different ways for each project.

Again, MER does not stray from the path of straying from the path (sic). For us, differences really come out full tilt when you sit in a different chair, literally and figuratively. On Tuesday of this last week, I began my TUL training. (Read: I began my campaign of world dominance.) It's been on my list of goals for some time now, but it's been one of those goals that I kept forgetting was one of my goals. Suddenly, there I was, in lockstep with and shadowing the prime TUL of the day. Suddenly, there I was, feeling like I did the first few times shadowing TAP/SIEs a year ago.

Mr. Prime TUL Man did his best to keep it high level and speak to the TAP/SIE within me (he's a qualified TAP/SIE, too, of course): "Are you on the Ninja Spotting email list? You need to be. You have to know where the ninjas are."

"Oh. Ok," I said. A little lacking in reception, I think. Although I was only 10 feet from the TAP/SIE chair, I didn't feel like it.

"What about the Space Godzilla analysis? Do you know where to find that?" Mr. Prime TUL man continued with the questions.

"No, no idea." I continued with the lack of answers.

It became very clear — nearly right away — that I'd have to start the mind-numbing process of learning where the TUL's bodies are buried. The TAP/SIE's job was already intricately detailed — a million places to hide bodies — and now I had to do it all over again, and for a job with more responsibility. 

As a TAP/SIE, you sit in your corner, poking your head into various conversations almost as you please. You perk up at the walkthroughs and meetings, but as soon as they finish, you inch away from the center of the controlled discussion going on at the round table and you go on your way. You have the power to pick and choose what you want to hear and when you want to hear it. Then, you can turn back around and keep turning the crank. You can ignore things. You can let the overwhelming rock of Riverside drown out the noise in the background, if you want. (By the way, you do want. They are the droids you're looking for.) 

As a TUL, the eyes are now on you. You can't ignore the room, the noise, the questions — except maybe when you go get your lunch. As I shadowed Mr. Prime TUL Man this week, I watched in awe of how many questions must be answered by the TUL. 

"What's the drop dead time?"
"How much margin did you give me?"
"What's the duration of the Pancam observations?"
"What's the expected downlink data volume?"
"How comfortable are you with the power situation?"

TAP/SIEs can blithely accept (with some reservation) that the TUL is going to take care of business. I certainly adopted this paradigm — you know, the "don't worry, that's TUL's problem, they'll take care of it" kind of paradigm. Now, I'm the TUL. (Er, the TUL Shadow. For now. Rome wasn't built conquered in a day.) Now, those questions are pointed at me. Now, everyone's looking at me. The job — nay, art — of the TUL is to take all of these questions and problems and turn them into manageable, single-bite cookies of Mars'y awesomeness. There's no chaos when there's a TUL in the room. And there's always a TUL in the room.

Now, to begin my world domination, I require sustenance

(Yes, there will be a TUL "album of the day" when I go prime in a few months. Yes, I expect my co-workers to Google the TUL AotD. I'm looking at you, Dr. Awesome XVII.)

Saturday, October 1, 2011

Rocks mean stuff, and stuff means science

Since I last wrote an update post, things have been calm in the MER world. I got lucky and landed three shifts in a row this week… right when things got pretty boring. As is well known by the avid MER fan base, MER has been brushing, RAT'ing, APXS'ing, re-brushing, and MB'ing at the Salisbury-1 target on the Chester Lake outcrop, in addition to capturing high-resolution Pancam mosaics of the Endeavour Crater:

(Thanks to Stu fo the stretch!)

When I arrived back on the tactical shift scene, we had started MB integrations. Which means not driving. Which means not moving the IDD. Which means shifts end earlier. Which means… a lot of things. The item of interest here is the strategic plan of using the MB. It is an instrument best described as a "dim light bulb," since its radiation source has gone through several half lives and longer integrations (collections of data) are required to achieve the same signal-to-noise ratio that was achieved during prime mission. The idea is that we must place the MB and stick around for at least several sols, and we can only integrate for data during warmer parts of the day. We have to balance this with infamous deep sleep behavior, remote sensing requests, and lack of coffee. (Always a problem.)

Chester Lake is, interestingly, a much softer rock than the science team expected. The RAT had no problem grinding right on through the outer layer, making a hole about 3mm deep on the first shot. This is where we left things off last time. What has happened since then? A lot, even though we haven't moved.

Due to a string of unrelated bad luck, we had to use the Runout sequence capability for sols 2724 and 2725. No biggie. This is how we live and breathe, baby. Turns out that space is hard. All we get are (useful) Pancam Taus. On sol 2727, we re-brushed the Salisbury-1 target. Why? The science team was unsure about the content of the extra tailings/rocky bits left over from the RAT grind a week earlier. Although the RAT brush is very worn, they felt that a re-brush would clear things up when we decided to…

Place the Mossbauer and start collecting data on sol 2728! The decision was made to stick around here because, hey, you know, we already made that RAT hole, and we'd be stupid to abandon it now. You hungry data miners will quickly find, however, that another apparent technical hiccup left us going into the Runout for this sol, so the plan got trashed in the end. (I spoke too soon.) Again, no problem here. 

Oh, hey, yeah, let's look at the RAT and a false-color of the RAT hole while we're at it:

MB-centric planning weeks are more low-effort to us because if we left the MB down at the end of the last plan and we don't want to move, we don't need any RPs in the room. This is, really, two sides of the same coin: it's good because then meetings get shorter; it's bad because we don't get fancy animations. (Oh wait! Also the joy of the presence of the RPs! How dare I forget! Although… they might be robots… so… I'm reserving judgement.) We we're in Nominal planning shifts this week, meaning every day we get to plug back in a give it another go. In some sense this is not ideal for a MB campaign. If we were in Restricted planning mode, we could get two or three MB integrations out of a single planning day, resulting in a better results-to-effort ratio.

(An aside: By minor coincidence, the famous Rudolf Ludwig Mossbauer passed away recently at the ripe age of 82. It's on our lien list to get a target named after him for his contributions to science. Per the LTP report: The Nobel Prize in Physics 1961 was divided equally between Robert Hofstadter "for his pioneering studies of electron scattering in atomic nuclei and for his thereby achieved discoveries concerning the structure of nucleons" and Rudolf Ludwig Mossbauer "for his researches concerning the resonance absorption of gamma radiation and his discovery in this connection of the effect which his name.)

So, what do these rocks tell us? What do they mean? Well, they mean stuff. And stuff means science. Trouble is, I'm no geologist. When a geologist uses words like "breccia," "fluvial-aluvial," and "volcaniclastic," I'm just as lost as someone listening to me use words like "space," "is," and "pretty awesome." Our geologists are trying to tell a story, remember. This story is multi-faceted and requires the entire suite of MER instruments to be told. We've captured some nice layering when we MI'd the hell out of Tisdale-2, and we'll have captured some 66 hours of MB data on Salisbury-1 by the end of the weekend. Where's the next part of… the stuff? The science?

Although there is a laundry list of targets that we'd like to visit, the next goal is Shoemaker Ridge:

It was a two-fold reason for landing on this decision. First, Shoemaker Ridge looks rich in targets and appears to contain a lot of… stuff. (Science.) Second, it's got north-facing slopes. Winter fast approaches — winter solstice being March 31, 2012, which is not very far away, and our ability to drive and use the IDD will fall steadily until then.

There were many other places we could have chosen next:

-What about ditching Cape York altogether and going south to Sutherland Point and Cape Tribulation? There's lots of stuff (science) in Botany Bay, which stands between Opportunity and Sutherland Point. 
-But why abandon Cape York, whose stuff (science) we haven't even fully characterized? We'd have to leave as soon as we were done at Chester Lake to get there in time to tilt our solar arrays on the 10-20 degree slopes that we'd find at Sutherland Point. 
-What about heading further north and finding some juicy Noachian on Cape York? There are lots of north-facing "lily pads" on the north side of Cape York, but it'd take some seriously unnecessary heroics to get that far north in a mere 100 sols or so.

There's some serious engineering going on here, but the choice is easy: There's stuff (science) to be done, and it's right under our noses. Shoemaker is where it's at, for now. All told, we've got ourselves a nice strategic plan. Shoemaker ridge will give us a nice view of much of Cape York and nearby north-facing lily pads. The choice was easy. No uncertainty or hesitation. Which there ain't a lot of in MER world. Uncertainty lies only on the tactical timeline — days — and not the strategic timelines — weeks and months. Any apprehensions fall by the wayside when the data from yestersol comes rolling in. Shift gears, take a breath, refill the coffee, and soldier on.

Then, more stuff. (Science.)


This weeks' albums of the day:

Sol 2728: III — Chickenfoot ("They're going to spend a lot of time looking for 2.") Here, take this totally egregious and contrived rock video. Take it. It's yours.
Sol 2730: This Is Spinal Tap — Spinal Tap (Self-referentialism! Again!)