Navigation in Space

[Whiskey144]

So, this actually came up in the Avatar MilWanker thread, where I asked what kind of navigation an ISV uses to get around. It turned out to be something similar to an idea I had, which as it so happens is a fairly universal way to get around the heavens.

Go figure I'd think of it, it's a common concept. Anyways, here's my question:

Would 3-axis stellar navigation make starmaps a useless commodity? Since, in a pinch, you can navigate with passive optics, a slide rule, and some trig tables, would a starmap be an obsolescent tool for practical use?

[Simon_Jester]

...Sigh.

Hardly. For one, using such crude tools would be a desperation solution- the possibility of making errors that will cost you dearly in reaction mass is high. For another, you need to know roughly where to look in order to find 'landmark' stellar features with the naked eye- on Earth it's fairly easy because we know north, south, east and west on the ground, can easily determine the plane of the ecliptic, can use known constellations of fixed shape as our guides, and can rely on the brightness relationships of the stars to remain constant.

If we're doing STL interstellar navigation at low speeds, we can still do this fairly effectively- but we need starcharts, which will have to be updated as we go to adjust for the fact that constellations don't look the same after you move five or ten light years. And the journey will take a LONG time. If we're doing STL interstellar navigation at noticeably relativistic speeds (like the starships in Avatar), relativistic red-shifting and blue-shifting, along with the effects of length contraction, will make judging the starfield and finding 'landmark' reference points by eye nearly impossible.

Bluntly, you need computer support in that case, or a small army of genius mathematicians who work very very hard and will get you killed if they screw up. And either way they will be referring back to known lists of the positions, motions, brightness, and spectral profile of stars, because that is how they know which landmarks they're looking at. You take your sightings on, for example, the stars Sol, Rigel, and Sirius; to do this you need to know where in the sky to look for them. Computers could do an exhaustive sky search very quickly and find the stars in question given even a rough clue as to where you are relative to those stars, then update the position of the stars in your sky continuously. For human beings, each individual 'point' sighting will be a laborious process, made far far worse if your starship is travelling at high STL speed.

If you were desperate, if it was a matter of life and death, and if you had plenty of magic torch drive fuel to play with, you could do it by sextant, pen, and paper. But I don't want you thinking that this is a wise way to do it if you have an alternative, or that it is an easy way, or that it somehow makes knowing the position and other properties of the stars around you "obsolete."

Come to think of it, I begin to wonder what the hell the question means- what do you mean by a "starmap?"

[Skgoa]

From the way I understand the OP a starmap is just what you would expect it to be: a 2d representation of the night sky as seen from earth. I.e. a projection of a set of known landmarks (including additional info like brightness and spektrum) onto a sphere that has Earth at it's center. (Lets forget about Earth's motion around the sun for a moment. )
Thing is: thats all you need. If the distance from Sol for every star etc. is given in the map, their actual position is computable. This CAN easyly be done with a pen and a piece of paper. You don't need to be a mathematician for that. Then its pretty easy for a trained user to navigate via sextant. You won't waste too much fuel, although constantly rechecking your course and making minute adjustments is required.
Now, letting a computer do all that and having your starmap in it as a database is much more effective. But who knows, maybe a set of back up charts for a significant number of points along your intended path of travel might actually be a prudent thing to prepare when compiling your flight plan. (Loss of power/computer/telescopes...) It should be trivial to have your navigational comp compute and print them out for you.

[Whiskey144]

f you were desperate, if it was a matter of life and death, and if you had plenty of magic torch drive fuel to play with, you could do it by sextant, pen, and paper. But I don't want you thinking that this is a wise way to do it if you have an alternative, or that it is an easy way, or that it somehow makes knowing the position and other properties of the stars around you "obsolete."

To be sure, I wasn't in the least suggesting that the method I described as "in a pinch" would be something to use unless absolutely necessary. My thinking was more along the lines of the skill be useful if you happened to have some kind of navigation system failure.

Come to think of it, I begin to wonder what the hell the question means- what do you mean by a "starmap?"

Well, the gist of the original question was "would 3-axis stellar navigation make a 3D map and/or star database an obsolete navigational implement", which is actually answered in your post; you still need to know where the reference stars are, as you said, and for that a star database or 3D map/representation would be necessary.

Even if interstellar transit is conducted with jump gates, you'd still need some kind of graphical map that shows the "links" between the "nodes" (systems).

From the way I understand the OP a starmap is just what you would expect it to be: a 2d representation of the night sky as seen from earth. I.e. a projection of a set of known landmarks (including additional info like brightness and spektrum) onto a sphere that has Earth at it's center. (Lets forget about Earth's motion around the sun for a moment. )
Thing is: thats all you need. If the distance from Sol for every star etc. is given in the map, their actual position is computable. This CAN easyly be done with a pen and a piece of paper. You don't need to be a mathematician for that. Then its pretty easy for a trained user to navigate via sextant. You won't waste too much fuel, although constantly rechecking your course and making minute adjustments is required.
Now, letting a computer do all that and having your starmap in it as a database is much more effective. But who knows, maybe a set of back up charts for a significant number of points along your intended path of travel might actually be a prudent thing to prepare when compiling your flight plan. (Loss of power/computer/telescopes...) It should be trivial to have your navigational comp compute and print them out for you.

Interesting...... Though what I was thinking of when I said "starmap" was more along the lines of some kind of 3D-database representation.

But you do bring up an interesting idea; having backup hard-copies of maps and charts that detail your flightplan or just the general space you're traveling might be something that a starship crew would carry.

[Simon_Jester]

I think I'd like to try a series of posts on how to do this in detail- or at least the theory; I am not an astronomer and am not a specialist with the kind of instrumentation or software you'd want for this job on an actual starship.

Begin with the question of picking the reference stars.

To determine my position in 3D space, I need to know my position relative to at least three fixed points- by "fixed" I mean "of known position relative to one another." It's OK that the stars themselves are moving (they are), as long as I know at any given time how fast they move and how far apart they are from each other.

Let me pick three reference stars. For choice, these stars should be bright, so that they can be seen from a very long distance; they should be distinctive, so that they can easily be picked out from the background. They should be close enough that you can find them easily, with no possibility of error, but far enough away that they will be separated by fairly large angles over most of the region you might choose to navigate- you're in trouble if all of your reference stars are blotted out behind the same nebula!

Incidentally, for this reason I should pick a number of backup stars, which meet the same qualifications as the others- I must have many choices, because I want to be sure of being able to take sightings on three stars no matter what happens.

But above all else, these stars' position must be known very very accurately. Our measurements of the distance to remote stars is (today) uncertain by hundreds of light years; if we had to start doing deep-space navigation with a fast FTL drive that could cause a lot of problems. Fortunately, one of the first things you can do with advanced space flight is send off the equivalent of the Hubble Space Telescope on a fast engine, move it a light-month or two away, and take a comprehensive picture of the sky... which gives you a long parallax baseline and lets you measure the distance to distant stars much more accurately. If we had an observatory somewhere out in the Oort Cloud, we could work out the distance to distant bright stars like Rigel to within a small fraction of a percent without much trouble.


Now, if you are travelling between planets in our solar system, this is easy- pick any two stars on opposite sides of the sky, then a third somewhere else not near the direct line between them. Keep picking more until you get exhausted or run out of highly visible stars in the sky.

If you are travelling across distances of a few light years it is still easy- pick any three reasonably bright stars in the general vicinity; Sirius is an excellent candidate if you are navigating the stars immediately around Sol. Sol itself is not a great choice, because it isn't really all that bright, though it will certainly work well enough for travel to immediate stellar neighbors.

If you are traveling over distances of hundreds of light years, you would be well advised to pick very bright stars, such as red and blue supergiants. Rigel is a good choice, as it will be one of the brightest stars in the sky anywhere within hundreds of light years of Earth; it worked during the Age of Sail and it can work for you.

Many of the other good candidates for navigation near Earth also have longstanding names (Antares, Betelgeuse, Deneb), because the entire point of the exercise is to pick exceptionally bright and outstanding stars. Not coincidentally, these stars are highly visible to naked-eye observation from Earth, and received unique names from the ancients.

Use your own judgment and the criteria above to decide what other stars to use. Again, you want a long list; as you move around you will find nebulae obscuring your line of sight to some of your choices. Also, do yourself a favor, don't pick stars with high variability (some of which would otherwise be pretty good choices).

[Eternal_Freedom]

To preface, I am an astronomy student, and as far as I can tell Simon has got it bang on.

In simple terms, working out your position in interstellar space is no different than computing your position on Earth from the stars, the scales are just different.

And as Simon said, we need to know distances much more exactly than we currently do, but that's doable if we have this FTL or fast-STL drive. Plonk a telescope in the Oort cloud (which is now thought to extend about 1/3 the distance to Proxima Centauri, a good light year at least), and that gives you a 1 light year baseline for parallax measurements.

A rought guess tells me that having a 2 light year baseline (one telescope on either side of the Oort cloud) would allow accurate measurements out to maybe 10,000 light years, which is a good portion of the galaxy. Of course, if you can put telescopes in other solar systems you could have even more accurate distances.

Incidentally, parallax is not the only way we can measure distances to stars. There is a nifty equation called the distance modulus, that will give you distance to a star if you know both it's apparent and absolute magnitude. For those who do not know what those are:

Apparent magnitude is how bright a star appears from Earth. The sun is -26, Vega is 0, and so on, with lower (or negative) numbers being brightest. Blame the babylonians for that, and for making it a logarythmic scale as well

Absolute magnitude is how bright a star is from 10 parsecs (or 33 light years). On this scale the Sun is about 4.

As far as I am aware this equation is more accurate at long distances than parallax is, but you need to know the absolute and apparent magnitudes.

As for pickign stars, as Simon said you want them to be bright, non-variable and separated by wide angles. So chosing the three stars in Orion's belt would not help, but chosing say Polaris (in Ursa Major), Sirius (in Canus Major) and Vega (in Lyrae) would work, as there are all separated by at least 70 degrees (the pole star and two stars on opposite sides of the celestial equator).

Phew, info-dump over. Hope that helps!

[Simon_Jester]

Next step: now that you are lost in space, you must find the reference stars you chose before setting out on your journey. Depending on how lost you are, this step can be either very easy, or a massive undertaking.

Ideally, your reference stars are among the brightest things in the sky throughout the region of interest. Unless you have been completely and utterly stranded in some remote corner of space by Act of Q, you will still be able to find them easily.

If you have a powerful computer and some moderately sophisticated telescopes, you can save a lot of time by using the computer to do a sky search- scan the telescope across the entire sky, identify the brightest objects, and that narrows down your list of possible reference stars- noting that some of those "brightest objects" will actually be relatively boring stars that happen to be very close by.

If you're doing it the old fashioned way, this will take a little longer, because it will be harder for you to discard boring but nearby stars with your own eyes through a telescope. An easy way to partially orient yourself is to identify the plane of the Milky Way (which will look pretty much like it would from Earth, at least in general- a band of distant stars whose light fuzzes out into a broad glow). Knowing roughly where the galactic plane lies gives you many clues to the location of your reference stars- if you can find one of them, you will have little trouble predicting where the other two are or ought to be.

The first step of orientation is incredibly important when doing things the old fashioned way because otherwise you cannot begin to guess where to look for a given star. If you have to search the entire sky to find Rigel, you're going to be at it all night because you'll have to exhaustively look at every bright star in the sky and go "nope, not Rigel."


Which brings us to the next step- conclusively identifying a star. If you have reason to expect the star to be the brightest thing in a known patch of sky, that's not hard: you look for a really bright star in that patch and yep, that's it. Realistically this will be all you have to do for routine sightings as long as you are not totally lost, or have somehow jumped out through a wormhole into an unknown region of space. If all you're trying to do is confirm that you have traveled 0.010 light years since you took your last star sighting rather than 0.009 or 0.011, that'll do the job.

If, however, you don't know where the hell you are (unlikely barring magic FTL), or your ship has spun or tumbled unpredictably so that your angular orientation relative to the sky is unknown (more likely)... then you will need something a bit more conclusive. Again, in this case, orienting against the Milky Way will help you do it the old-fashioned way; if the Milky Way surrounds your ship like a big donut with you in the middle of the hole, it is a pretty fair bet that the nose of your ship is pointed out of the plane of the galaxy. There are only two directions that fit that description, which narrows down the area you need to search for your reference stars a lot.

With a decent infrared telescope you can also locate the general direction of the galactic core, which makes things even easier- but at that point your toolkit is sophisticated enough that you should damn well have a decent computer to be doing this for you. With passive visual instruments alone, well, no such luck.

Anyway, the problem of positively ID'ing a star when you aren't sure it's the right one is a little tricky. The best way I know of to do it is by spectral analysis: any star has a unique spectral fingerprint in the form of certain absorption lines, colors for which it radiates very little light of that wavelength. Playing around with a diffraction grating, some very precise angular measurement devices, and a notebook can get you this; I did it with elemental spectra off a mercury vapor lamp as a lab assignment once.

An instrument to do this job cannot be improvised, at least not without one hell of a machine shop; if you want to be able to do this, make sure to pack the right toolkit before leaving home.


Now, if you are standing still relative to the star, that gives you the star's spectrum. If you are moving at a noticeable percentage of c (and even a few percent is noticeable, if you're doing this precisely), the star's spectrum will be red-shifted (if you are moving away) or blue-shifted (if you are moving towards it). The degree of red or blue shift varies relative to the angle between the line to the star and your line of flight; stars that lie at "3 o'clock" or "9 o'clock" relative to your line of motion will hardly be shifted at all. You can correct for red-shift and blue-shift as long as you know how fast you are going, but it adds a step of complexity, and in extreme cases the shifting may move some of a star's spectral lines out of the visible spectrum. Which is, of course, not a problem if you have the right instruments to observe infrared and ultraviolet light... but if you can do that, you should damn well have brought a computer to automate this process.

Otherwise, you're doing calculations on the degree of red/blue shift by hand, and adjusting the observed spectral profiles of your bright star accordingly to see if it matches the profile of your reference star.


Eventually, you will have identified three reference stars and their positions in the sky relative to your ship. This may involve a great deal of complicated mucking about with a telescope, a diffraction grating, and an observation blister, plus assorted reams of paperwork... or it may involve sitting down with a good book while your computer does all the work.

Now that you have identified those stars, all you need to do is measure the angular separation between them. At that point, you can finish the job and calculate a position "relative to the fixed stars" (actually, the slow-moving stars whose current positions are plotted or calculable from your log-books), as an exercise in spherical trigonometry which is left to the reader.

This can be done with a maritime sextant if you have to, though the typical sextant is precise only to something over a tenth of an arc-minute, which you can't count on to give you the angle to precision better than, oh, about 0.01%. Under many conditions you'd like to do better than that, because if your reference stars are hundreds of light years away, that level of precision means an uncertainty in your position of many, many billions of kilometers.

Today, more precise instruments exist or can be designed by competent engineers; go ask them for details.


Incidentally, if you've done this procedure to sight on distant stars, and you have a detailed map of the stars immediately around you, you can do the procedure again on the nearby stars to get a more accurate positional fix using your limited-precision toolkit. You get more accurate results from measuring the relative angle between stars 10 light years away (like Sirius, if you're near Sol) than between stars 1000 light years away (like Rigel, if you're near Sol).


There. Did I miss anything important anyone can see?

[Simon_Jester]

Addendum: if you suspect that you are traveling at relativistic speeds and don't know how fast you are going, well. My first instinct is to advise "put your head between your knees and kiss your butt goodbye."

But if you're determined and well-prepared, hmm. Wait! I know what you can do!

Get a directional microwave antenna, tune it to around 160 GHz. Point it in front of you and listen for a suspicious hiss. That hiss is the cosmic microwave background radiation, which is nice and pretty much uniform throughout the universe, so you can count on being able to pick it up.

If you do not hear a suspicious hiss, check to see if your microwave antenna is broken. If it's not broken, something is wrong with the universe... relative to your frame of reference. Far and away the most likely explanation is that you are (as suspected) going very fast.

Keep turning up the dial to higher frequencies until you start hearing that hiss. That is the cosmic microwave background again. Record the frequency at which the hiss is strongest, to the limit of the precision of your hardware. Compare to the known 160.4 GHz peak frequency you would expect to see if you were at rest "relative to the fixed stars." Do a quick calculation of relativistic Doppler shift and you will have your speed, to the limit of the precision of your hardware.*

Just to be extra-careful, as it pays to be when you are traveling at relativistic speeds, point the microwave antenna behind you and repeat the process, this time shifting down from ~160 GHz to find the peak intensity of the cosmic microwave background radiation. Now do the Doppler shift calculation again, and hopefully you should get the same answer.*

At this point, you are officially permitted to cry out to an uncaring universe:

Science. It works, bitches!
____________

*Incidentally, modern astronomers have already done this to high precision with our own galaxy to measure its speed.

[Knife]

So, this actually came up in the Avatar MilWanker thread, where I asked what kind of navigation an ISV uses to get around. It turned out to be something similar to an idea I had, which as it so happens is a fairly universal way to get around the heavens.

Go figure I'd think of it, it's a common concept. Anyways, here's my question:

Would 3-axis stellar navigation make starmaps a useless commodity? Since, in a pinch, you can navigate with passive optics, a slide rule, and some trig tables, would a starmap be an obsolescent tool for practical use?

Simon is all over this, but I thought I'd toss my two cents in.

I would imagine a scenario where as the ship's computer has a map/model of the local stars and their predictive motions; including planets and all known navigational hazards like comets and asteroids. The ship would chart it's course via this model, then as a back up, periodically checking via passive or active sensors, depending on the occasion, the location of the actual stars/planets/planetoids to check it's position against the chart.

[GrandMasterTerwynn]

Next step: now that you are lost in space, you must find the reference stars you chose before setting out on your journey. Depending on how lost you are, this step can be either very easy, or a massive undertaking.

Ideally, your reference stars are among the brightest things in the sky throughout the region of interest. Unless you have been completely and utterly stranded in some remote corner of space by Act of Q, you will still be able to find them easily.

If you have a powerful computer and some moderately sophisticated telescopes, you can save a lot of time by using the computer to do a sky search- scan the telescope across the entire sky, identify the brightest objects, and that narrows down your list of possible reference stars- noting that some of those "brightest objects" will actually be relatively boring stars that happen to be very close by.

Your navigational computer should be loaded with a database of a specific kind of highly-luminous star called a Classical Cepheid variable star. They're useful in determining distances because the relationship between their period and luminosity is well-understood. There are dimmer Type II Cepheids and RR Lyrae variable stars which help us establish distances to globular clusters and the like.

There's also the method of locating yourself that the Pioneer 10 and 11 probes have engraved on their plaques (for purposes of telling anyone who finds them where Sol System is,) and that's by looking at pulsars; since we've quantified their rotational periods to high levels of precision.

If you're doing it the old fashioned way, this will take a little longer, because it will be harder for you to discard boring but nearby stars with your own eyes through a telescope. An easy way to partially orient yourself is to identify the plane of the Milky Way (which will look pretty much like it would from Earth, at least in general- a band of distant stars whose light fuzzes out into a broad glow). Knowing roughly where the galactic plane lies gives you many clues to the location of your reference stars- if you can find one of them, you will have little trouble predicting where the other two are or ought to be.

On the topic of orienting yourself with respect to the Galactic disc, you can look for globular clusters, whose sizes and star populations are pretty well-understood. You can also look for the Small and Large Magellanic Clouds and the Andromeda Galaxy, as those are big obvious objects.

[Simon_Jester]

Your navigational computer should be loaded with a database of a specific kind of highly-luminous star called a Classical Cepheid variable star. They're useful in determining distances because the relationship between their period and luminosity is well-understood. There are dimmer Type II Cepheids and RR Lyrae variable stars which help us establish distances to globular clusters and the like.

You don't really need to know the luminosity of the star, though that helps a bit.

Luminosity just tells you how far you are from the star, and that's not necessary information. If you know the distance relationships among your reference stars (Betelgeuse is X light years from Rigel, which is Y light years from Deneb, which is Z light years from Betelgeuse, and they form the following triangle), that's good enough. All you need is the angles to those stars to fix your position, not their absolute distance.

Knowing distances helps you locate yourself if you only have one or two reference stars to work with... but in that case you can't fix your position with 100% confidence anyway.

There's also the method of locating yourself that the Pioneer 10 and 11 probes have engraved on their plaques (for purposes of telling anyone who finds them where Sol System is,) and that's by looking at pulsars; since we've quantified their rotational periods to high levels of precision.

Pulsars are tricky. Unless I am much, much mistaken, they only provide a detectable signal along a more or less fixed cone that their beam sweeps out; if you're not in the right place you can't see a pulsar at all.

Blue giants, on the other hand, are omnidirectional beacons- as long as you don't get too far away for them to be visible they'll always steer you correctly. For purposes of doing navigation the old-fashioned way (God forbid) that would be vital. Even with computers, it's handy.

On the topic of orienting yourself with respect to the Galactic disc, you can look for globular clusters, whose sizes and star populations are pretty well-understood. You can also look for the Small and Large Magellanic Clouds and the Andromeda Galaxy, as those are big obvious objects.

Yes, that works well too. Reference stars are good because they're concentrated point sources- you can take very very precise angular sightings off them and get an exact position that way. Using clouds, nebulae, or globular clusters gives you rough positions well, but isn't so good for precise navigation.

[GrandMasterTerwynn]

Your navigational computer should be loaded with a database of a specific kind of highly-luminous star called a Classical Cepheid variable star. They're useful in determining distances because the relationship between their period and luminosity is well-understood. There are dimmer Type II Cepheids and RR Lyrae variable stars which help us establish distances to globular clusters and the like.

You don't really need to know the luminosity of the star, though that helps a bit.

Luminosity just tells you how far you are from the star, and that's not necessary information. If you know the distance relationships among your reference stars (Betelgeuse is X light years from Rigel, which is Y light years from Deneb, which is Z light years from Betelgeuse, and they form the following triangle), that's good enough. All you need is the angles to those stars to fix your position, not their absolute distance.

Well, yes, but the period/luminosity relationship makes this easier to do with these variable stars. If you know how far you are from a couple big, bright, variable stars; and you know how far they are from wherever your home port is (or, indeed, from each other,) and where they are with respect to 'home,' you'll pretty much know exactly where you are.

There's also the method of locating yourself that the Pioneer 10 and 11 probes have engraved on their plaques (for purposes of telling anyone who finds them where Sol System is,) and that's by looking at pulsars; since we've quantified their rotational periods to high levels of precision.

Pulsars are tricky. Unless I am much, much mistaken, they only provide a detectable signal along a more or less fixed cone that their beam sweeps out; if you're not in the right place you can't see a pulsar at all.

Unless an act of Space Poseidon hurtles you completely out of charted space, there ought to be ephemeris detailing which pulsars are visible in what parts of the charted space. It's still a good backup system to fixing yourself with respect to bright stars with well-quantified identifying behaviors.

[Ahriman238]

Since far wiser heads have this well in hand, all I can contribut is a gross oversimplification:
The tools for ships to navigate by star and sun are centuries old, the technology to do so by satellite coordinates is at least a decade old. However, these things did not invalidate the existence of charts. Rather, they made the existing charts all the more valuable, since you could now draw a dot on the map and say "we are here. To get where we want to go, we should turn 15 degrees north in the next few hours." If you're willing to dig around for it, there's actually quite a story in how they got started measuring longitude, though I doubt you could apply it to space travel.

Pulsars are tricky. Unless I am much, much mistaken, they only provide a detectable signal along a more or less fixed cone that their beam sweeps out; if you're not in the right place you can't see a pulsar at all.

Correct me if I'm wrong, astronomy hobbyist here, but didn't the SETI plaques on the various out-of-system probes contain a starmap based on pulsars to show earth's location?

Blue giants, on the other hand, are omnidirectional beacons- as long as you don't get too far away for them to be visible they'll always steer you correctly. For purposes of doing navigation the old-fashioned way (God forbid) that would be vital. Even with computers, it's handy.

Rare enough to serve as good landmarks too.

A question: I believe we have a fairly good fix on the stars of our galaxy. Would it not make sense to divide the galaxy into smaller quadrants with three or four well known landmarks that can be seen from anywhere inside the quadrant? Then include that data with every ship, if you're at all concerned about getting lost? "Hmm, a blue giant, three pulsars and an unusual nebula, we must be in 16A! Okay everyone, earth is that way but Beta Colony is a lot closer in the opposite direction."

[Skgoa]

Simon is right on most of this, so just a few minor points:

- No, limiting yourself to a few reference points is NEVER a better idea than having a full star database. You want to make as many meassurements as often as possible, to increase precission... and other objects might obscure some of your landmarks. You also don't want your data in the form of "X has Y distance to Z", because that just unneccessarily obfuscates information. Yes, you can store the most important information on your reference points in a somewhat smaller space... but you buy it with a much higher computational complexity to get anything useful out of that.

- Actual navigation will not be limited to using the stars. The most important tools will be inertial guidance and computations based on the maneuvers that have been done. I.e. the navigational computers will have an internal model of where they think the ship is and what it's course is. Meassurement of reference points will be a backup tool that is used to increase precission on normal flights. Your computer will make thousands of meassurements every second, but your primary source of knowledge about your whereabouts is the knowledge of where you have been. I would think you will also get a very precisse meassurement of your position and heading from every STC (space traffic controller ) and every other ship you encounter.

[Simon_Jester]

Pulsars are tricky. Unless I am much, much mistaken, they only provide a detectable signal along a more or less fixed cone that their beam sweeps out; if you're not in the right place you can't see a pulsar at all.

Correct me if I'm wrong, astronomy hobbyist here, but didn't the SETI plaques on the various out-of-system probes contain a starmap based on pulsars to show earth's location?

They did, but that doesn't make them the best navigation system. The advantage of using pulsars is that they are very easy for interstellar aliens (basically the only kind that could conceivably intercept the probe) to locate, even from extremely long distances, and that no matter how long the probe is out there, those pulsars will still be around- for many thousands, for millions of years at least, and it will take that long for the Pioneer and Voyager probes to get very far over interstellar distances.

But pulsars are not equally visible from all points in space. As navigation landmarks, they leave something to be desired.

Blue giants, on the other hand, are omnidirectional beacons- as long as you don't get too far away for them to be visible they'll always steer you correctly. For purposes of doing navigation the old-fashioned way (God forbid) that would be vital. Even with computers, it's handy.

Rare enough to serve as good landmarks too.

Rarity helps, but brightness is important.

A question: I believe we have a fairly good fix on the stars of our galaxy. Would it not make sense to divide the galaxy into smaller quadrants with three or four well known landmarks that can be seen from anywhere inside the quadrant? Then include that data with every ship, if you're at all concerned about getting lost? "Hmm, a blue giant, three pulsars and an unusual nebula, we must be in 16A! Okay everyone, earth is that way but Beta Colony is a lot closer in the opposite direction."

That would be a routine practice, but to answer the question fully, no, we really only have good information on visible-light profiles of stars within the nearest several thousand light years, with a few exceptions. One of the advantages of pulsars is that they are visible from very long distances, that is true. But for observation that works anywhere, using relatively primitive instruments, there is a lot to be said for relying on blue giants.

Nebulae are a bad choice because their shape will vary depending on your viewing angle; pulsars are good, blue giants are good. You might well define "sectors" some thousands of light years across, which you can orient yourself in by observing galactic-scale features like the Magellanic Clouds and the galactic core, then list the reference stars that will be useful within that volume (again, Rigel will be useful anywhere within a few thousand light-years of Sol, and for a volume a few thousand light-years across over on the other side of Rigel from us).

Simon is right on most of this, so just a few minor points:

- No, limiting yourself to a few reference points is NEVER a better idea than having a full star database. You want to make as many meassurements as often as possible, to increase precission... and other objects might obscure some of your landmarks. You also don't want your data in the form of "X has Y distance to Z", because that just unneccessarily obfuscates information. Yes, you can store the most important information on your reference points in a somewhat smaller space... but you buy it with a much higher computational complexity to get anything useful out of that.

In the passage you quoted, I was talking about doing this with little or no computer support- in which case you cannot take a hundred star sightings 'just in case,' unless you are very fortunate or numerous or have a lot of time to play with. If you have good computer support, as I have repeatedly said you should, this is trivial; the computer can do a hundred star sightings in short order and nail down position even faster. It is good to do a hundred star sightings. It is, at a stark desperate minimum, necessary to do three.

I am trying to come up with a method that illustrates the basic steps involved (find star in your reference table, identify it by its spectrum, check against positions of stars in reference table to calculate position) and that can (in theory) be done using limited, crude instrumentation.

As to the format of data stored on reference stars, my point is that the minimum information you need to know is the relative position of your reference stars, however many of them there are. You do not, strictly speaking, need to know the position of the red dwarf two light years off your ventral bow. You do not, strictly speaking, need to know the exact position of the galactic core except insofar as the positions of your reference stars are defined relative to the galactic core (which they might be, in some settings). You do not, strictly speaking, need to know the exact distance to your reference stars- if the angular positions of those stars is measured, and the spatial relationships among those stars is known, you can calculate for yourself how far away you are from each of them as a fairly simple matter of trigonometry- i.e., something a man can do with pen and paper, though it will take him a long time. And it is far, far easier to measure the angular positions of stars than the luminosity of stars.

You do need to know the distance between the reference stars- that Rigel is X light years from Deneb and so on. That is not all you need to know, and it is certainly not all you want to know. But it is enough to permit you to, in principle, compute things like "how far from those stars am I?"

- Actual navigation will not be limited to using the stars. The most important tools will be inertial guidance and computations based on the maneuvers that have been done. I.e. the navigational computers will have an internal model of where they think the ship is and what it's course is. Meassurement of reference points will be a backup tool that is used to increase precission on normal flights. Your computer will make thousands of meassurements every second, but your primary source of knowledge about your whereabouts is the knowledge of where you have been. I would think you will also get a very precisse meassurement of your position and heading from every STC (space traffic controller ) and every other ship you encounter.

Yes. Since this discussion originated because someone was asking "how to do this using primitive instruments," I was focused on coming up with a roughly viable method that could, in principle, be done using passive optical instruments and pen and paper. It is not the best method. The best method is to have a decent suite of instrumentation hooked up to your starship's computer, which will no doubt cost a trivial fraction of the value of the ship, and rely very heavily on that computer for star sightings, inertial guidance, and everything else.

But if you had to do it with a telescope, a sextant, and a diffraction grating, that's pretty much how you'd do it.

[Guardsman Bass]

What a fascinating thread. I hope it gets saved, like the "Sensors" thread.

[Ahriman238]

What a fascinating thread. I hope it gets saved, like the "Sensors" thread.

Seconded.

[someone_else]

Rarity helps, but brightness is important.

What about Hypergiants?
Rare and bright. Yeah, they last just a few million years, but it's still a buttload of time for humans.

Of course those are going to be the things you look for in case Q threw you around the Milky Way, for everyday navigation there are much better stars.

Or not? I'm no expert. Just google-fu-ing.

[Edi]

What a fascinating thread. I hope it gets saved, like the "Sensors" thread.

*kicks*