Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Not so much a thought experiment as a single neuron misfire.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.

user:
pass:
register,


                   

supercomputer to space

a project to displace human beings in space for conducting science experiments in LEO
  (+2)
(+2)
  [vote for,
against]

human beings need water air light food poop medicine for the occasional sickness and, to an extent company ( at least one person to keep them company) .

in space this means they need air filter systems water filter systems sanitation systems lighting medicine and first aid equipment enough of these for at least 2 people ( so they dont go crazy)

all the talk lately of supercomputing and AI soon reaching a threshold of human capability means that we should seriously be considering how to PRICE the difference between launching the weight of a fully capable AI system into LEO space that can displace 2 people and all the systems they need to live in order to carry out their mission to conduct science experiments.

to begin with we must accept a big assumption---that this AI brain and body we are discussing is capable of replicating the thought process and physical process of astronauts as they complete their varied missions in space. that is a big assumption of course, but i'm putting it out there so it does NOT become a point of contention.

now the question is how many KILOGRAMS of AI brains and brawn do we need to replace 2 people on a future dedicated mission where 2 people and their dedicated life support systems would be totally replaced by a robot inside a chamber that is large enough to accomplish science experiments in air. note that the robot can also likely accomplish experiments on the outside of the station in the vaccume of space as necessary, but for now we are assuming the air itself is pure nitrogen in a robot only science space station.

well, ignoring the fact that 2 people sleep ( assume future AI needs to sleep just like humans) then a future AI will displace

2 humans (160kg) their clothing bedding and space suits ( 100kg) their dyhrdated food for a short mission of 1 month on future space stations ( 40kg ) their WATER for 30 days ( 40kg) their water filtration and recycling system ( 80kg) their physical bathroom and the solid waste it stores ( 160kg) the air ( ? let's assume the robot operates in the same partial pressure environment as astronauts -------but with all oxygen swapped with nitrogen to eliminate fire risk)

the fire supression system no longer necessary in an all nitrogen space station( 10kg ) oxygen refiltering system no longer neceesary ( 50kg)

a return module is no longer necessary to bring the crew back down ( hundreds of kilograms but this does NOT need to be included as a return module may already be waiting in outer space for astronauts Launched to a station or with their own statino)

my thoughts here are that you can launch a 640 kg supercomputer with 8 hands capable of experiments and motion around a station that would displace 2 astronauts and all their life sustaining gear. power systems and requirements are assumed to stay the same meaning that the robot benefits from all the excess electricity that would otherwise be use on life support systems.

more likely, the true weight of is much higher, see the links below for the weight of the soyuz modules.

realistically, a 2000 kg pound super computer robot ( 4000+ pounds) would be economically feasible to replace astronauts when it is CAPABLE enough of doing science experiments .

this means as robots as smart as astronauts ,get smarter and lighter than 4000 pounds they will displace astronauts in outerspace. NOTE THAT I AM ASSUMING THE ROBOT BEHAVES AUTONOMOUSLY BUT UNDER INSTRUCTION LIKE CURRENT ASTRONAUTS

the robots capability for remote operation in near real time may actually make the robot as capable as astronauts at conducting science experiments without actually needing to be smarter than them; so current AI might be sufficient to achieve this challenge with sufficient bandwidth)

the point of this exercise is for you to help me come up with a kilogram assements of the actually launch weight at which unmanned systems that are assumed to be as capable as astronauts will displace astronauts due to cost and risk concerns for astronauts health.

side points. no Public relations risk. robot can conduct experiments in conditions not amenable to humans ( on the outside of the stations in the space vaccum) robots can work 24/7 robots cannot mutiny they cannot die as a person, but only have problems that can be fixed or not. a modular ai can isolate and quarantine unfixeable problems in its cuircitry. so risk of malfunction is lower than humans.

their missions can be extended or cut short at will while they are placed into dormancy or let to burn up upon re- entry.food water and air constraints are eliminated, making missions MUCH more flexible

massive COSTS of life support monitoring , health monitoring, food preparation, etc....are avoided.

robots have none of the air/water/medical/sanity/boredom risks of astronauts.

and of course astronauts can be sent up to FIX the system if it breaks.

my point here is that man's time in space in coming to end , at least temporarily , as our AI technology development supersedes our ability to cheaply launch very heavy objects into LEO that carry human beings for the purpose of science experiments.

soyuz 3 is 7000 kilograms with 8.5 cubic meters for habitation.

teslaberry, Jul 21 2016

soyuz https://en.wikipedi.../Soyuz_(spacecraft)
[teslaberry, Jul 21 2016]

soyuz 3 http://spaceflight1...ecraft/soyuz-tma-m/
7000 kg and 8.5 cubmic meters of space [teslaberry, Jul 21 2016]

cubic meters of space compared for conducting science experiemtn http://globalstorag...storage-calculator/
[teslaberry, Jul 21 2016]

Baked & WKTE for decades https://en.wikipedi.../Robotic_spacecraft
[notexactly, Jul 21 2016]

[link]






       Yes. You can learn a lot from unmanned probes, but they are not exploration. It aint explored until somebody's been there.
MaxwellBuchanan, Jul 22 2016
  

       So, a reusable redeployable spacecraft is what I'm getting here. On-board AI navigates to various celestial bodies, carries out observations, banks the data and moves on to the next. Deimos fly-by, search for life in the Delta Quadrant, swing by the sun to recharge, etc. [+]
whatrock, Jul 22 2016
  

       I do think it would be interesting to send missions of paired robots, to see what they could do that a single couldn't. Belaying each other on climbs and traverses. Performing troubleshooting and maintenance on each other. Taking wide-displacement stereo-grams. Being the "for scale" object in each other's pictures. And, of course, getting the Pulitzer prize winning pictures of when the other 'bot bought it...
lurch, Jul 23 2016
  

       Seven megagrams isn't a lot for a supercomputer, especially one that has to be radiation resistant. And by the time one is ready to fly newer hardware will have the same performance at half the weight.
Voice, Jul 24 2016
  

       But we must go in case some alien wag puts a Pokémon on the far side of Neptune.   

       Reusable deployable is heavier than single-use specialized because it must be bulletproof for all kinds of conditions. Although if you got the formula right it could be cheaper in the long run, but I doubt it, since it is highly probable that it would still become specialized for each mission, thus taking the worst of both approaches.
RayfordSteele, Jul 24 2016
  
      
[annotate]
  


 

back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle