anghara (anghara) wrote,
anghara
anghara

Launchpad Day 6, AM - Space, the final frontier... for humans and computers...

We had two talks this morning.

The first was by visiting lecturer Ruben Gamboa, on the subject of computing in astronomy.


We started out with the basics - computers in general are useful in controlling equipment, automating repetitive tasks, organizing and retrieving data, building and exploring scientific models. We looked at a few of those in context - in the astronomy field, controlling equipment might mean things like telescope control (we saw some of that last night up on the roof), allowing us to do things which would otherwise be difficult and tedious in terms of controlling tracking, etc, and vehicle control (for example, high-tech planes are flown almost entirely by computer and not manual manipulation of any actual, you know, CONTROLS in terms of hands-on-by-human. But the basis for much of this is old technology which is rooted in the world of sensors (which sense the environment or context) and actuators (which then act on it). SOme examples of these would be telerobotics (like waldos) and unmanned vehicles which are still nevertheless remotely directed by a human operator from a distance.

The latest technology would be the systems powering the Mars Rovers - they may have limited memory (which works just fine for space missions standards under the context) and their computing power is equivalent to last-generation PCs (amazing to think that your old desktop could run a ROver on another planet, isn't it?). But they are still dependent on human intervention. For instance, the Rovers cannot function during the Martian night because it is too cold - so during the night of inactivity human "drivers" decide on the Rover's next mission and then they will build a sequence of commands, simulate them locally to make sure that what will happen is what they want to happen, and send them to the Rovers. They are literally writing custom programs every day and sending them to an alien planet - where one misstep could be expensive and/or fatal. Bugs in the code could be a pricey oops. There are still lots of "what if" human decisions - but the next generation of Rovers could be closer to true AI because they will have commands about what to do and precisely how the task is accomplished will be left to the Rover AI on the ground on Mars. (By the way, an interesting tidbit was that when the original Mars mission went out with the first Rovers, the mission had to launch during a particular launch window to get to where they needed to be when they needed to be there... but the software wasn't ready on time. No problem... they *uploaded it later*. The mind boggles...)

Computers are useful for automating tasks that are
- too important to be left to (careless) humans
- too boring for (most) humans
- too time intensive

For instance - searching for comets.

Old school comet hungers scan the sky, comparing what they see with a memorized mental map. This is pretty hard to do well, but many amateur astronomers enjoy it. It’s a competitive “sport” – takes 400 of observation hours to have a chance of spotting a comet. And the stakes are high – discover a comet, get handed immortality with your name in the heavens.

Automated searches – perfect candidate for automation. The computer stores a “mental” map of the sky and “looks” at that region every day, determining if anything changed (moved) or a “new” fuzzy object appears.

The same method can be used to look for supernovae. Traditional approach is to “blink” two photographic plates of the same location, to see any visual differences. Computers can do this continuously.

But there is also something to be said for humaning tasks - some tasks, even boring ones, are still better left in human hands. The Galaxy Zoo, for example, uses human power (distributed via the internet) to classify galaxies. There is real science here, relating to the evolution of galaxies.

Data processing is what made computers popular and affordable. Its roots go back to accounting and stuff ugh

Astronomical data – observation data today are made in digital form. Telescope data is captured and stored in computers, and then processed there. Large scale data can cover a wide range of sky in different wavelengths - for instance the Sloane digital sky survey (SDSS). Collected data is made available to the public and - its more complex form - to professionals in the field (professional access includes spectra, photometric data, spectroscopic data) through the Sky Server it includes tutorials and other educational materials and also hosts fantastic images. It's public access.

Search facilities are vital – lots of room for research here – data hosted in a relational database users can search the data using frontend tools or providing raw SQL commands
SQL is ok but it’s from the seventies but we really want easier ways of searching the data – SQL interface is hard

Astronomy with Google – the next generation of survey telescopes will push the limits of data storage – the large synoptic survey telescope LSST) 8.4m telescope, 10 square degree FOV, 30 terabytes of NEW data PER NIGHT – that's 30 000 gigabytes – or 40 000 CDs – or almost ½ of the Library of Congress – or 1/20th of the totality of material posted on YouTube. This is mindboggling. Google is partnering up with LSST to provide IT support for data processing, much as Microsoft partnered with SDSS – not Google’s first foray into scientific data – already volunteered to host terabytes of scientific data and to make it available to other scientists and to the public – data can be donated by sneakernet (send a hard drive for it to be actually physically downloaded by someone).

Page rank – original idea – better web pages have lots of links to them – but the real problem isn’t so much finding matches as prioritizing the matches.

Google goes to academe – the idea is that scientific computing can take advantage of the same platform that Google uses to index the web – the term is cloud computing

The new overlord – backing from Google, IBM and NSF – cluster exploratory – 1600 processors, terabytes of memory petabytes of disk, projected use includes modeling of the brain – not there yet but getting there.

The advent of the computer to this degree has led a number of people to postulate that this is the end of science as we know it - from here on, it's all DATA. But the ideal sequence of events in developing a scientific idea would be Tycho Brahe (data collection)--> Johannes Kepler (looks at data and sees patterns – don't know why true but it happens and we can make a case for it, or see it)--> Isaac Newton (looks at data and at patterns and says, I can EXPLAIN that).

Today we use computer models – the solution is to integrate models by simulating them in a computer this lets us view results predicted by a model even if it can’t be solved directly.

Computer technology had a lot to offer to astronomy and space science – remote exploration, data management, data processing. Computers can also provide new insight and new techniques to advance science.


Jerry Oltion returned to present the second session of the morning, on humans in space.



What actually happens to a person in space – what happens if there’s tear in your space suit? Well, you aren’t going to explode (unlike in the movies) – you might asphyxiate when you lose oxygen but you won’t explode – if you lose pressure you have about 15 seconds. You will swell up slowly to about 2 human sizes as gases come out of your bloodstream – people tend to desribe it as cold but not accurate or at least not instantaneously there is a psychological component. Someone exposed to this for almost a minute was subsequently revived and put in a hyperbaric chamber and he was literally fine after that.

Here's some info on space suits.

Can you see the stars from space? Misconception always see a dark sky full of stars but if your eyes are adapted to normal light (e.g through a window from a well lit space station) you will look at a view from such a window and see black sky – if you peer right through the window shielding your eyes you will see stars – but just like on earth you need to be in a darkened environment to see the stars.

Most stars will look white – even a red K type star will look white to your eye

Living in space – what is it going to require – well – everything. We need air, fluid, shelter…

Recirculating/regenerating air supply – you have to find a way to make the air smell good – nose will eventually stop responding to bad smells but your mind does not and you will always come back to the bad smells, especially after you’ve just woken up – chemical scrubbers but not wholly successful – rubber gaskets and stuff outgassing, people’s bodily odour – it all accumulates. You have to make the FOOD taste good – you can get a little stuffed up because of physiological reasons and that actually makes things taste bland – it might be different for kids born in space.

The pure disorientation of zero gee and if you grab someone else to stop yourself spinning you will set BOTH of you spinning – Ender’s game battle room

What are long term effects? After several generations born in space will be lose the need to orient ourselves the same way as the guy next to us in a two-dimensional plane.

Air in space doesn’t naturally convect – it’s a gravity driven phenomenon. Hot air rises cold air flows – doesn’t happen. So you need to fan-cool every heat generating source by hand which makes it noisy (“like standing next to a diesel locomotive”, space shuttle astronaut) so at some point it makes more sense to spin space station to get artificial gravity rather than go the fan way because it would drive everyone crazy.

Sounds terrible, doesn't it? So why go there in the first place?

If just want to see if can live in sealed environment, you can do it here on earth too and it’s safer in case of catastrophic failure. For those that go “because it is there” – well, you’d need to have a lot of disposable income. Can take an existing asteroid and burrow into it and seal it and you have a “Spaceship” – Larry Niven’s idea – asteroid with iron nickel, hollow it out put water in turn sunlight onto it wait until it all melts to the water and then boils out and if time it right freezes at roughly spherical – and you have a hollow space station…

Nobody’s talking about sex in space – because it possibly isn’t all that good – pressure issues might make it, ahem, difficult for the men (“to become alert”, according to Mary Robinette Kowal). Also, things don’t just float around in space they start spinning around a central mass of gravity – and if you grab someone you increase angular momentum – but if your spins are at odds you can twist people around and there[s no faster way to get someone throw up which kind of kills the romantic impulse right there – and then once you get it together you kind of drift apart so, Velcro suit…? As far we know it has not been done successfully in the wild.

But you need gravity for fetal development to occur properly and experiments have not turned out happy – we are designed for gravity and things just don[t work correctly in zero g – it might be that we might have access to drugs or other medical intervention but right now it is a problem.

Using astronomy in fiction - Stanley Schmidt says "tell them that it’s unlikely to have a habitable planet around stars that have names" – because stars are easily visible, and have been so for long enough to have recognisable names, because they are really really hot or else they have gone into red giant stage and incinerated everything in their path.

What’s actually visible when - important in terms of worldbuilding. You can see some constellations in summer but not in winter and vice versa - Orion up in winter in northern hemisphere. Everything you see rising today will rise 4 minutes earlier tomorrow – which is why we have a whole new zodiacal constellation in the sky every month.

The moon of course is different – 28 days to go around the earth and rises roughly an hour later every day – full moon always straight up at midnight (except for daylight savings…)

Sidereal day – one earth rotation from and to a star – always the same – stars don’t move wrt ourselves. 23 hours 56 minutes
Solar day 24 hours
When you add it all up there is one secret “day” in a year…

If moon rising in the sky tide coming in if setting the tide is going out – always lags just a little behind the moon but in general it’s true. Multiple moons may have complex and potentially cataclysmic effects on tides, both on fluid and on continental mass which also responds to gravitational pull. Both can profoundly affect the lives of the people who live on the planet which that moon or moons is/are orbiting. DOn't ignore these human details when writing a story about things that are larger than life.


That's the morning. The final afternoon sessions - see next post...
Tags: launchpad
Subscribe
  • Post a new comment

    Error

    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded 

  • 0 comments