perogiepro
perogiepro
99 / 13
24th Aug 2015
22nd Dec 2015
We are the SFPI, we are searching for mechanical life that can detect danger and think. essentialy basic AI. Our end goal is to make one that can replicate and change, so it will """evolve"""
machine intelligence experiment neuron life survival challange evolution spfi

Comments

  • msasterisk
    msasterisk
    25th Aug 2015
    Oh, and if you need a green light scanner I can retrofit my PSTN robot's eyes to do that.
  • msasterisk
    msasterisk
    25th Aug 2015
    Oh wait... I didn't read your comment, perogiepro, so I only just realized that if they are cells of one body, evolution is indeed possible! On an unrelated note, I find the notion of a disembodied brain remotely controling an organism rather amusing... ;)
  • msasterisk
    msasterisk
    25th Aug 2015
    Because if we do that, then any mutation will surely kill it, and thus it is incapable of evolving. Or are we trying to make an AI? Or both? Just sayin...
  • perogiepro
    perogiepro
    25th Aug 2015
    and the neural thing is good for keeping the bot randomly moving and detecting and going. not for intellegent thought. and the bot breaks lesss now but i need to make it perfect at that first
  • perogiepro
    perogiepro
    25th Aug 2015
    that's promising! i put a neuron network in there temporallily while i work on the new brain. About the replicater. If they all share a brain then maybe we can treat the bots as CELLS of ONE organism. They need green light sensors and such but im just going to add the replicating abilty to this bot.
  • msasterisk
    msasterisk
    25th Aug 2015
    Wait... we're not using a "messy bot" design, instead we are trying to keep it intact?
  • Technomancer
    Technomancer
    25th Aug 2015
    As far as not needing neural networks, you are quite correct. I'm still working with the CRAY brain and hope to have something publishable later today, but basically what I'm doing is generating a mass of (pseudo)random signals, then filtering them so only the useful ones are acted upon. It's almost the opposite of how organic brains work, but I'm having good results so far.
  • Technomancer
    Technomancer
    25th Aug 2015
    Have you actually looked at Devas' replicator? It's at least 4-5 times as big as the robot here, even without any mobility or awareness. And if you make one of these things copy itself, both will have to run from the same brain, and both will feed sensory input INTO the same brain. I'm not saying it can't work, but it's going to be bloody hard to stop them all getting confused and having a mass breakdown.
  • perogiepro
    perogiepro
    25th Aug 2015
    and the bot can reproduce after reaching plant (food) so we should implement devastheempty's replicating bot design into it.
  • perogiepro
    perogiepro
    25th Aug 2015
    ok we dont need a complex processor or neural network or such. I think we just need to make it a brain that just takes sensor inputs and applys them to the correct movements. such as a photon sensor that only lets green light in (filt) so when it dectects photons that bonce off plant, it moves towards it.