Eugene Leitl wrote:
> On Thu, 22 Mar 2001, Michael Lorrey wrote:
> > Okay, here is what we really need bad to make micro-sats and microprobes
> > feasible: an SSPS: Solar System Positioning System so that space probes
> SPS, rather.
> Yeah, it's overdue. Actually, since matter tends to aggregate in
> recursive/hierarchical arrangements, you need to retain several
> positioning system (me-relative (ultrasonics and solid state camera),
> Earth-relative (GPS), Sun-relative (SPS), Galaxy-relative (er, UPS, I
> guess)), and use coordinate transforms from one another.
Excellent. Actually, you'd need to with MPS, for galaxy, since our
galaxy is the Milky Way and galaxies tend to be named M####. Then CPS
for Cluster Positioning System, then UPS.
> Actually, with IPv6 you can combine the logical and physical addresses,
> e.g. by shifting WGS 84 latitude/longitude/altitude to 32 bit each,
> reserving the leading 32 bit of an IPv6 address for future purposes.
> If I get my math right, that's about 1 cm resolution at Earth surface (so
> it's not enough for nano, blame the IPv6 designers. If one thinks of it,
> there are lots more of addressable objects in the local galaxy that
> can be described by 32 bit either. Clearly we shouldn't have settled
> with mere 128 bit but would have gone straight to 256 bits).
Well, we clearly need to do so from an encryption standpoint as well.
Don't want any aliens finding their way around the solar system too
easily, they might land a cruise photon torpedo on us.
> > can easily figure out their position at all times and run their
> > navigation routines off that data rather than having to deal with star
> > sighting, wasting fuel and/or power on spinning the spacecraft around
> > trying to get a reading. I mean, what are we, in the 17th century or
> I don't know, wouldn't a few low-res solid state cameras would give you
> plenty on orientation relatively to stars, and more than a bit of
> navigation when used with planets, if equipped with a good clock? Given
> that we now have 3-axis accelerometers with better than 1e-6 g resoltion,
> there is probably not all that much requirements for a solar positioning
> system (SPS). Plus, with isotropic radiators you'll need a lot of wattage
> to cover the terrain, unless you stud the space evenly with nodes each
> covering a space of a lightsecond, or so.
I'm primarily interested in simplicity, so cameras and so forth are too
much. Each face of the cubic bus will have a photosensor head for rough
intensity sensing to determine the direction of the sun. Everything else
will come from SPS transmissions of Satellite transmitter ID, timestamp,
and its own known position. Beyond the three E-S lagrange sites, an SPS
transmitter should be put in orbit around each planet as a planetary
> What probes really need are built-in survival reflexes. A little orrery in
> their digital brain, which plots their future course, and some dumb AI
> which knows "it is probably not a good idea to fly into solid objects at
> Mach 19", "account for estimated future burns in respect to fuel use",
> etc. This will make them refuse obviously suicidal instructions, unless
> explicitly overriden.
A very good point. Tell it to go point A to point B and let it figure
out how to get there.
> > what? Why are we still using ancient navigation techniques? There should
> Because we're not a spacefaring species, and only poke our noses a little
> outside of the atmosphere.
Sure, but smoke signals are old hat too.
> > be an SSPS satellite in each of the three Earth-Sol Lagrange points.
> > This would be a good investment of public money, and its even
> > constitutional for libertarians: space is a 'channel of commerce', or a
> > 'navigable waterway', so the gummint is responsible for maintaining
> > their navigability.
> > I am currently designing a 180x200x200 millimeter probe bus that will
> > use GaAs/GaAnt solar cells fed focused sunlight from 2 inflated 2 meter
> > dia. mirrors (theoretical output: 2.99 kW) to power the electric
> > thruster of choice. Fuel tank is 150mm dia spherical. Pick your own wall
> Maybe you can help me:
> When thinking about a cloud of coke-can sized satellites for digital pulse
> radio and wireless networking (no stabilization, realtime beamforming
> using active antennas), one has the following problems: the systems need
> to be gossamer, yet able to survival rough (shrapnel-size) deployments.
> Initially I thought about using vanilla rectangular monocrystalline Si
> cells on springed hinges (jack-in-the-box type panel unfolding). However,
> these things are rather brittle. So I thought freezing the sat into a
> sublimating solid cube (water, or dry ice), which will vanish within
> few hours under orbital conditions. However, this will eliminate the
> advantages due to higher weight. I then thought of a photolythic foam
> envelope, then of a pyrotechnical envelope (shrapnel deployment scenari),
> which doubles as protection and propulsion.
> Any ideas in that area?
Aerogel. How cold can they get? I'd say frozen hydrogen, but thats
conductive as well as corrosive.
I think that ice is the lightest you could possibly get unless you use
some foam concept.
I have a similar problem with an idea I have to suspend hydrogen and
oxygen in a solid matrix, to create a hydrogen oxygen rocket that uses a
solid fuel rocket's simplicity of design.
> > thickness, pressure, contents. Maneuvering thrusters will be solid state
> > pulse ignition chemical micro-thrusters off-the-shelf. Navigation will
> > depend on SSPS to be in place.
> > You decide what payload you want and how quickly you want it wherever
> > you are going. I figure with this size bus you could deliver a laptop
> > sized replicator facility to any NEA within a few months of launch.
> Officer, arrest that man. He's obviously dangerous.
This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:59:42 MDT