Tag Archives: Kerberos

P–15 days: Navigating Pluto

As of today, Monday June 29th, I find myself airborne headed for the Applied Physics Laboratory in Maryland while 13 hours earlier, at midnight leading into this morning on the east coast, New Horizons found itself 18.5 million kilometers from Pluto and moving 1.2 million kilometers closer every day.

The Dreaded Target-plane Drift has come and gone. For about a week, the solutions drifted downward and leftward on the target plane—that big dart board in the sky—each time more tracking data was added, until finally it departed the bottom of the acceptable target box, a 300 x 200 kilometer rectangle centered about 14,000 kilometers down (Ecliptic south) and left (eastward) of Pluto.

The cause of the Dreaded Target-plane Drift is not yet completely understood, although it’s almost certainly related to earlier errors in locating the centers of Pluto and Charon, and this was most likely caused by inadequate knowledge of the albedos of those bodies, the dark and light patterns covering their surfaces. It’s hard to model the surface brightness of a body you’ve never seen up close before.

The DTpD, while it was in play, was a very worrisome thing because the Navigators didn’t know if it was real or—much worse—caused by a hidden problem, maybe serious dynamic mis-modeling that’s been overlooked to this point or even an unknown bug in the software. This caused some very restless activity and sleepless nights for several days in a row. It now appears that the drift has stopped and the current position, just outside the box, is correct and the earlier positions were not. There was probably not a dynamic mis-modeling problem or software bug. The moving finger of Navigation writes, and having writ, moves on.

In any event, the Dreaded Target-plane Drift may be a thing of the past as the Navigators rely more on the pin-prick images of Styx, Nix, Kerberos, and Hydra. Those bodies are small enough that no appreciable diameters are yet seen and the albedos of their surfaces doesn’t even enter the problem. So … while the DTpD may never rear its ugly head again, the Navigation Team will not hold its collective breath. Eternal vigilance is the norm.

At this red hot moment, if the current dart in the dart board is the correct one, surrounded by it’s little cloud of Gaussian-shaped probability, then the science observations won’t be completely optimal at closest-approach time, and the intended occultation of the Sun and Earth by Charon may not happen.

Thus in the wee hours of tomorrow morning the penultimate maneuver, TCM17B1, will execute and move the drifted solution back into the box, the happy return of the prodigal son. It will cost a few centimeters per second change in velocity, and then all will be well.

We hope.

P–21 days: Navigating Pluto

New_Horizons_-_Logo2_big

Continuing the countdown, focusing on optical navigation using images of Pluto and the satellites.

As of midnight this morning of June 23rd, Eastern time, we’re 25.4 million kilometers from Pluto and moving 1.2 million kilometers closer every day.

The whole thing comes in a rush; time is not a leisurely independent quantity here—it’s relentlessly urgent! Another maneuver approaches fast: TCM17B1 executes in seven days, the next-to-last opportunity to correct errors before the flyby, and Navigation needs to come up with the “correct” trajectory ahead of that to support its design and implementation, a process that will take most of that seven days.

They need answers now! Both Navigation Teams—PNav (Project Nav) and INav (Independent Nav)—are busting their butts to figure out what’s going to happen in the target plane—that big Plutonian dart board in the sky. The error ellipses shrink down slowly as we get closer, but the solutions wander around, some of them inside the previous error ellipses, some out, depending on what combinations of optical data are used and a host of other assumptions, all in the form of experiments to figure out the most realistic answer. How many toad tongues and bat wings do we need to add for a brew that is true? (With apologies to Danny Kaye and The Court Jester; I couldn’t resist.)

Ideally new solutions wander around inside all the older ellipses as the newer ones shrink down around them. Lately INav has been holding fairly steady, but PNav dithers around, sometimes in, sometimes out of the ellipses computed only a day or so earlier. The game right now is to figure out which OpNav (Optical Navigation) modeling is correct, and which misleading. Pluto itself seems to be the problem.

Images of Pluto, that is. Since we’ve never seen this Honorary Planet up close before, extracting an accurate estimate of its center from fuzzy optical data is as much art as science at the moment, and nobody will know what the real answer was until we actually get there and know all the Plutonian “blemishes” in detail; the craters and icy plains and mountains and scarps, and whatever else we find.

However, at this red-hot-moment, they are only just beginning to pop out in sharper detail, the dark regions and the light, and in order to realistically extract the center of a 2400 kilometer wide body to within a few tens of kilometers, you have to model the blemishes with an accurate “albedo map.” The problem is, the details pop out faster than you can model them, so you play a catch-up game with the Pluto images, and if the current albedo map is behind the times, so is your estimate of where we’re

Plutonian Albedo Map

Plutonian Albedo Map

going in the target plane. There are other models for center-finding too, and they’re also in the mix of experiments, all of them throws-of-the-dart to see where we come up relative to Pluto—all of them to see which ones we believe.

INav seems to have avoided the problem up to now by the simple expedient of not using Pluto images at all and relying on images of its companion, Charon. Because Charon is smaller, it has smaller center-finding errors. PNav is coming around to the same conclusion: Don’t trust Pluto!

Fortunately, the smaller satellites Nix and Hydra are beginning to pop out of the optical wood-works, and because they are little more than specs in the images, mere pin-pricks of light, they will be much better located against the starry sky. After a while tinier Styx and Kerberos should become usable too. Ironically, the best information of where we’re going relative to Pluto is going to come from those satellite pin-pricks in the images, and not from Pluto itself.

Meanwhile, I depart for Maryland Monday to become a small part of this Plutonian melee with whatever humble capacity I can bring—it’s been a long time since I’ve done this kind of interplanetary navigation, and the learning curve is extremely steep and precipitous.

And meanwhile, nobody can stop the relentless march of time as Pluto expands in our vision and the activities grow more urgent. The pressure is on.

May_12_Pluto

P–30 days: Navigating Pluto

 

Goldstone_DSN_antenna

Goldstone 70 meter antenna / NASA / Wikipedia

2015/06/14. Sunday
The third of a series of posts about navigating to Pluto

Distance: About 36 million kilometers from Pluto at 5:30 am Arizona time as I write.
Velocity: A not-much-changing 13.8 kilometers per second toward Pluto and away from Home. Not-much-changing because the pull of Sun’s gravity isn’t very strong out here.

I think I’ll do nuts and bolts today: how do the Navigators do what they do? That means I need to talk about the DSN (Deep Space Network) and navigation data types and other technical what-not. Let’s try to keep it light and entertaining, shall we? If you’re not technically inclined, you may be excused from today’s session, but be warned you will be tested on the material in the final exam. Attendance at the next session is mandatory.

But first, News Flash! The maneuver this morning was successful as I see in an email from Alice, the mission operations leader: “Initial TCM assessments are showing a nominal burn within the expected parameters.” Chris, one of our Navigators at the Mission Operations Center at APL (Applied Physics Laboratory) in Maryland says “The Doppler residual came up right around 1.06 Hz … indicating very likely a nominal burn.” Woo-hoo! We’re walking a little under two feet per second slower toward Pluto so that we’ll get there about 83 seconds later, close to the intended arrival time.

Back to the Deep Space Network. What a romantic, adventurous, ambitious name! I can hardly believe it’s real; the term evokes wonder every time I give it a little thought. Especially since there wasn’t any Space when I was growing up, much less Deep Space. I mean, most people didn’t think we could ever go into Space, and if you thought differently you were a Space Cadet.

I was a Space Cadet. All the way up to Sputnik in 1957. Then Space became IN. I thought I’d become IN, too, from Space Cadet to Visionary in a single launch, but no, that didn’t happen. It’s hard to project the necessary gravitas at age fifteen. Since then, I’ve navigated spacecraft to every planet in the solar system. From zero to sixty in only seventy-two years! (And I’m still a Space Cadet. When does the gravitas kick in?)

Where was I? Oh yes, Deep Space: a term that never fails to evoke romance, mystery, and adventure. Are we earthlings actually tracking things in Deep Space? Yes, we are! And someday, if we’re good and don’t kill ourselves first, we may even go out there ourselves.

The Deep Space Network: There are three locations almost evenly spaced around the world where NASA operates antennas that track New Horizons. The locations ensure that we (New Horizons) will be in sight of at least one of them all the time. There are several antennas at Goldstone, California; several near Canberra, Australia; and about an equal number not far from Madrid, Spain. We call those sites Goldstone, Canberra, and Madrid for short, but more commonly we talk about the antennas themselves, like DSS-14 (Deep Space Station 14, a 70 meter wide antenna at Goldstone) and DSS-65 (one of the 35 meter antennas at Madrid.)

The DSN antennas collect four different tracking “data-types” from New Horizons and funnels them to the Navigators through various channels that aren’t important to this discussion. Here they are (bear with me, it gets a little thick for a few sentences here and there):

(1) Doppler data: A DSN station sends up a radio frequency signal. It travels 4.5 hours to the spacecraft. When the spacecraft gets that uplink, it sends back a downlink signal that’s in “harmony” with the uplink. That means that in the process of turning the signal around and retransmitting it, New Horizons accounts for the frequency of the uplink on a cycle-by-cycle basis. When it gets back to Earth 4.5 hours later, it’s compared to the frequency that went up. If it’s the same, then nothing has changed, and that’s evidence that we live in a static universe in which nothing moves! That doesn’t seem to be the case. We always see a shifted frequency, pretty firmly establishing that we live in a non-static universe!

eeEOoo

Doppler effect defined

The received signal is different from what went up. Why? Ta-da, the Doppler effect of course, named in honor of Christian, its discoverer. That’s the eee-ooo sound you hear when a train goes by, but radio waves behave the same. And when you analyze that shift, you get clues about how the spacecraft moves, and just as important, how the tracking station moves because of the earth’s rotation, and when you put all that information together over the course of a tracking pass that lasts several hours, you can figure out not only how fast New Horizons is moving away from the Earth, but also you can determine its location in the sky: the Right ascension and Declination (reverting to astronomy-speak for a moment). The Doppler data is incredibly powerful in navigation, at least in the radial direction, and a change in New Horizon’s radial velocity of only 1 millimeter per second will look like a big signal.

So there—we’re through the hardest part of the discussion I think.

But wait, there’s more!

(2) Ranging data: If you time the signal (and the DSN has extraordinarily good timers, accurate to a gnat’s ass (another technical term), and know the speed of light (which we do) you can get the distance to the spacecraft to an accuracy of much less than a kilometer out of 4.7 billion of them. Now there’s a truly astounding accuracy). There are a lot of complications, of course, but that pretty well sums up the big picture for ranging data.

But wait, there’s more!

(3) DDOR data: Sometimes we use two of the DSN stations simultaneously, like Goldstone and Canberra and—over the course of about an hour—alternately track the spacecraft and then a quasar near it in the sky. Quasars are conveniently loud at radio frequencies, and they’re so far away that they don’t budge over the course of many, many years,

Delta-DOR defined

Delta-DOR defined

so they provide a very nice fixed reference system (thanks, Mother Nature!) for figuring out the direction of the spacecraft to jaw-dropping accuracy, about one-millionth of a degree. This data type has the gawky-gangling name of Delta-Differential One-Way-Range, which we usually shorten by calling it Delta-DOR, or writing it “DDOR”.

But wait … (oh, never mind). There’s one more.

The three types above—Doppler, ranging, and DDOR—are so-called radiometric data types. They’re all “centered” at Earth, so-to-speak, so the farther away New Horizons gets, the less accurate they are. To add to the uncertainty, we don’t know the distance from Earth to Pluto very well yet, so even though we might know the distance from Earth to New Horizons to that gnat’s ass, we don’t know New Horizon’s distance to Pluto to better than, very roughly, 1000 kilometers. That leads us into:

(4) OpNav data: The 4th data type, optical navigation data, or OpNav in the vernacular, is based on pictures taken from New Horizons of the things out in front of it, namely Pluto and his retinue of satellites, downlinked to the DSN and thence to Navigation. The OpNav team, a subset of the larger Navigation team, is led by Coralie. She and her team pick the locations of the tiny blobs of Pluto, Charon, Nix, and Hydra out of the noise of the images (a very difficult and tedious process with all kinds of complications) and compare them to locations of stars in the same images. This “pins” the spacecraft down against the stellar background, and since the locations of the stars in the sky are well know, so is the location of New Horizons. (The remaining two known satellites, Styx and Kerberos, aren’t used for the OpNav process because they’re too small and hard to see.) The location of the stars, Pluto, and satellites in the images constitute the data passed to the Orbit Determination team led by Fred.

So, that’s the end of the data descriptions. Not so bad, eh, if you’re still with me.

Since the OpNav data is spacecraft “centered”, it gets more powerful as we get closer to Pluto, until finally there comes a time when it overwhelms the accuracy of the radiometric data and tells us where we’re located relative to Pluto rather than Earth. From then on, OpNav data is the prima donna of the navigation show except for complications (always complications!) like when the spacecraft gets really close to Pluto and the Doppler data begins to “feel” the gravity. This doesn’t happen until the last day in the flyby because Pluto is so NOT massive (at least as compared to the planets).

It’s the job of Fred and his minions (of which I may be one), to boil all of this big slurry of data down (remember that big black witches’ pot in the first post?), scads and scads of it (another technical term), until there’s nothing left but a tarry black residue. That’s what we call the solution. It tells us where we are and where we’re going.

But wait … It’s not cut and dried with just one solution. There are a lot of unknown parameters that go into the witches’ brew, and the Navigators have to make assumptions about their values and how much they trust those assumptions. There are also a host of other variables, too much to go into, so just call them toad tongues and minced-spiced bat wings. They all affect the solutions to varying degrees.

The Navigators boil down many, many pots of witches’ brew, stirring vigorously, tasting occasionally and adding different amounts of toad tongues and bat wings to suit, so there are finally a lot of those tarry residues at the end of the process of orbit determination, each with a different taste (read trajectory). It’s the Navigators’ job to assess the flavors of all these and decide which one meets the reality of a successful Plutonian encounter, then deliver that nugget of information—a predicted trajectory—to the other teams of the mission so they can act on it.

So there it is in a nutshell.

There is no more. Today.