[Cyborg] rat brain robots
mrericboyd at yahoo.com
Fri Oct 8 14:54:09 PDT 2010
Thanks for the paper. I too am pretty skeptical, although I do think
they are doing the right thing by having the electrodes there for the
neurons right from the beginning. Maybe they provide enough "structure"
that the neurons can figure something out?
They do say in one of the publicly accessible first pages that the hard
problem is getting the neural net to share your goals, i.e. say you want
it to avoid obstacles, how do you train for that?!?
Skory - do you know if there is any audio or video available from the
talk that you say at CMU? I'd love to listen to that!
On 10/8/10 12:26 AM, Todd Anderson wrote:
> The only paper I can retrieve in full is the ethics article (see
> attached). I haven't read the paper yet - so please forgive the Donny
> Doubting - but I do wonder about cultured neuron experiments in general:
> can we expect neurons thrown together without a developmental wiring
> process (which, for example, would usually lead to 3D connection maps)
> to produce a great amount of computational power? I am quite sure they
> make excellent artwork, as in Stephen Potters work with Symbiotica (at
> the University of Western Australia)!
> On Thu, Oct 7, 2010 at 6:37 AM, Eric Boyd <mrericboyd at yahoo.com
> <mailto:mrericboyd at yahoo.com>> wrote:
> Kevin Warwick has been busy! Apparently they have been culturing rat
> neurons, wiring them up to robots using electrode arrays and bluetooth,
> and figuring out how to train the "rat brain" to drive the robot around!
> Videos show progress as being pretty good!
> The article links three academic papers they have written, but of course
> the full text is behind paywalls. Anyone have access to those journals?
> Amazingly, Warwick's homepage (kevinwarwick.com
> <http://kevinwarwick.com>) doesn't mention this
> research at all, especially since there are articles from the BBC about
> this from 2 years ago.
> I also found some research by Steve Potter of GATech in 2002, although
> he seems to be focussed on the neural culturing and electrode
> technologies, rather than the cyborg aspects.
> Cyborg mailing list
> Cyborg at lists.noisebridge.net <mailto:Cyborg at lists.noisebridge.net>
More information about the Cyborg