NavList:

   

Reply

I saw some posts of links that might be of interest to the list.  Two
follow.  The second is a longer discussion of space navigation that
might be of interest.  In regard to space navigation, as best as I can
tell, they often plot positions on an earth-centered coordinate system,
much as we do, and get distance from the center of the earth by
ranging, such as timing the round trip of radio signals.

Fred

ONE

While we're seeing a burst of astronomical OS X packages getting a
mention here, I thought I'd add that the venerable astronomical
data reduction system, Figaro, was ported to OS X a while back.
(If anyone is really interested, it's on the AAO's anonymous ftp
site, ftp.aao.gov.au/pub/figaro/4.5.1.figaro.tar.gz or
4.5.1.macosx.exe.tar.gz for a pre-built version for Panther.)


TWO

>>> On March 10, around 12 pm, EST, Nathan Strange wrote:

>>> I noticed a couple of posts on the list by astronomers, and I thought
>>> I'd mention JPL's SPICE library which can be found at:
>>>
>>> ftp://naif.jpl.nasa.gov
>>>
>>> in:
>>>
>>> /pub/naif/toolkit
>>>
>>> and:
>>>
>>> /pub/naif/toolkit_docs
>>>
>>> This library is now widely used in NASA, ESA, and I think NASDA too.
>>>
>>> Not only does it let you read in SPK ephemeris files for the planets
>>>  and moons, but it has a lot of handy functions for transforming
>>> orbit
>>>  elements, calculation occultations, etc. And that same FTP site has
>>> spacecraft
>>>  ephemeris files that you can use to animate stuff like Cassini's
>>> arrival
>>>  at Saturn :)
>>>
>>>
>>> You can do all of the calculations done on the horizons web page
>>> , and more stuff too.
>>>
>>>
>>> As far as I know this is the most complete open-source astrodynamics
>>> library available....  but there's a few things I'd like to add
>>> like a lambert solver :)
>>>
>>>
>>> I've used the fortran version of the library on both linux and os x
>>> to solve some fun problems.
>>>
>>> Anyway, that's my plug for the library... I hope some of you find it
>>> useful...
>>>  (Any NAIF programmers on the list feel free to correct any errors
>>>  in this message)
>>>
>>>
>>
>
>

Nathan Strange
Cassini Navigation
Jet Propulsion Laboratory

(The opinions expressed in this email are the opinions
of Nathan Strange alone and do not necessarily represent
the official position of any organization including JPL,
NASA, and CalTech.)

>> On Mar 10, 2004, at 5:12 PM, Fred Hebard wrote:
>>
>>> Nathan,
>>>
>>> Would you mind if I copied your post below to a navigation list of
>>> which I'm a member (Nav-L).  I was going to do it without
>>> attribution to protect your privacy, but then thought it better to
>>> ask!
>>>
>>> Thanks,
>>>
>>> Fred

> On Mar 11, 2004, at 11:30 AM, Nathan Strange wrote:
>
>> No problem, feel free to forward it with my name if you'd like.
>>
>> Is it a sea-faring navigation list?  The library was primarily
>> designed for space-faring
>> navigation.  It would be neat if it was also useful for sea-faring
>> navigation.
>>
>> -Nathan Strange
>>
>>

On Mar 11, 2004, at 9:35 AM, Fred Hebard wrote:

> Hi Nathan,
>
> Thanks.  It's a sea-faring or air-faring navigation list, devoted to
> traditional celestial navigation.  I have no idea how precise you have
> to get for space-faring, but 0.1' of arc is plenty accurate with
> ephemerides used at sea with hand-held sextants; in small yachts and
> airplanes, 1.0' of arc is fine.  With GPS, celestial navigation is a
> dying art, but the Navy, Air Force and Merchant Marine still use
> celestial.  The Armed Forces like to have a non-emissive system of
> navigation; I believe all the Air Force systems are automated, but the
> Navy still uses hand-held sextants, as well as automated.  The
> automated systems can be quite accurate, from what I understand.
>
> There are a number of computer systems out there for calculating
> ephemerides and position at sea.  The Naval Observatory and the
> British put out a package, and they also still put out printed
> versions of the Air and Nautical Almanacs.  I just thought the list
> might be interested in what you had.
>

Sounds like some fun stuff!  My boss, the Cassini Nav Chief, is a fan
of Bowditch's, 'The American Practical Navigator'
and recently found an old edition.   When I get the time, I want to
take a celestial navigation class :)

JPL is the keeper of the most accurate ephemerides in the world.  The
most current right
now is de405 which is on that ftp site in my email.  de405s is a
smaller file that doesn't go as far into the future.

The horizons website would be the easiest way to access the ephemeris
data, and I think it would give you the
information that you need for celestial navigation.


> Some dumb questions.  Do you all use an earth-centered system of
> coordinates for near-earth operations?  If so, do you try to determine
> your distance from the earth using celestial observations, and how?
> Thanks again,
>
> Fred
>
>

I don't know much about near Earth navigation, but I can tell you about
how we do deep space...
We use a lot of coordinate systems, but mostly we use EME200
(Earth-Mean-Equator of 2000) because
it's the most standardized. We use a lot of different centers, so we
always quote distances with
stating what central body was used, and often what ephemeris was used
to locate the central body.
We use kilometers-kg-seconds units, but the rest of US aerospace
industry uses feet-pound mass-seconds
which is what caused the famous Mars climate orbiter mishap.

There's two parts to the navigation.  One part is generating the
ephemerides to predict the locations of the
planets and moons.  The other part is solving for the spacecraft
location relative to the Earth.

The big dish antennas in the California desert, at Madrid, and Canberra
track the spacecraft.  We get
angular data from the antennas as to where they found the spacecraft
signal.  But because of the distances
involved angular data isn't accurate enough.  We also use what's called
range data which is basically timing a
pulse sent from the Earth that reflected from the spacecraft back to
Earth.  Using the speed of light
this gives a very good distance measure.  They also measure the doppler
shift of the signal from
the spacecraft to get a velocity measurement.  There's another type of
data that I don't understand
called Delta-DOR... this is something involving using two ground
stations and a quasar for a frequency measurement
to get better doppler data.

All of this data is Earth based.... I guess this would be like the
control tower tracking an airplane
and radioing its position back.  But there are two more types of data
that use measurements from the spacecraft.
Just recently there have been enough spacecraft at Mars that we've been
able to use radio tracking between
spacecraft.  I think the mars rover landings were the first ones to do
this.   More common is what's
called optical navigation, and this is probably the closest analogue to
using a sextant.  For OpNavs the spacecraft
takes pictures of moons or planets and we use the location of the
objects in the picture (i.e. which pixel is the
center of the object) and the locations of objects relative to each
other.  OpNavs give us information on the
spacecraft location and on the ephemerides for the moons and planets.

And finally there are observations from astronomers around the world
and with Hubble that are used
to improve the ephemerides.

All of this data is collected a weighted and combined with models of
the forces on a spacecraft, and
of the gravitational fields from the planets and moons, etc...  and its
all crunched through
a program using a Kalman filter...  a Kalman filter is similar to a
least squares curve fit,
but it lets you weight some data as more precise than others and let
you add a model of the dynamics.

I think GPS devises use a Kalman filter, and I think Dennis Tito made
his fortune by applying
a Kalman filter to the stock market.

Sorry if this explanation is too involved,  JPL has a web page that
explains a lot of how we navigate:
   It might be easier to follow than my
ramblings.

I think for near Earth operations they mostly just use angular data and
doppler... but I'm not sure.

-Nathan

   

Reply