Robots Should Walk Like Penguins?
The Houston Chronicle has an article about how the walk
of Penguins could help build better robots. Robots that try to mimic a
human walking gate require cumbersome computers and expensive sensors to
keep them from falling over. A University
of Houston Professor, Max
Kurz, has found that the side-to-side waddling motion of Penguins may be
a mechanism that introduces stability and helps the little guys to
balance. This Penguin walking gate knowledge could help both robots and
humans. Robots could benefit from an inexpensive balance mechanism and
perhaps droves of elderly and stroke victims could learn to walk like
Penguins to assist in their balance too (that's a scary visual!). No other
animal no birds walk like Penguins do and we could learn a lot from the march of the Penguins. Similar articles...
stairs, posted 24 Jan 2006 at 15:45 UTC by JamesBruton »
(Master)
How to penguins go up stairs?... as far as I can remember they kind of
'hop up' onto things. I wonder if a machine that walks like a penguin
was as tall as the average human it would have sufficient stride to step
up onto a normal human stair.
With shorter legs it could certainly have a lower centre of gravity
(full of batteries)... sounds a bit like Robosapien v1 in all.
Isnt it counterproductive to make robots that walk like thiswhen they
are walking and running already? has the guy ever heard of P3 or
Asimo?
Perhaps this is tech going backward, but if you ain't got the money for
a Asimo, then you take what you can get. :-)
Not to devalue his work, but Jame Brutin can make a human sized biped
for less than Asimo then so can others.
whether he can do it, that is not what the article is about. It's about
making something cheaper. Not that I'd want some cheap thing, but if it
comes down to designing with 18 motors as compared to say like 6 motors
and choosing a different walking gate, then it might be a good
alternative for certain situations.
Wooooooow, crazy shooting Chris...
Several things then:
-In the video on my site Mr Stick Legs is walking (on two legs) with
only 3 wiper motors for the legs, he's about 5ft tall... getting a 2
motor upgrade at the moment but anyway, enough about that.
-I can understand the concept of making a simple walking gate, it being
cheaper and so on (because that's what I'm attempting myself). The only
issue I can see with using little penguin legs is the point I originally
raised about navigating human stairs and other obstacles in a human
environment.
-I thought the whole idea of making humanoid robots was to enable the
robot to fit in with humans... walking the walk... talking the talk...
otherwise just use wheels and tracks to go up stairs.
-Can't really comment on patients recovering from strokes... this is a
robotics site.
I wonder if penguins walk the way they do because it helps them from
falling over on the ice whereas longer legs would be easier to slip.
I'm sure it has a lot to do with the environment they live in.
adaptation, posted 30 Jan 2006 at 08:25 UTC by JamesBruton »
(Master)
I think it may be a matter of adaptation :-)
http://www.chron.com/disp/dispcomp2.mpl?cid=952606#_self
Apparently it learned in two weeks...
(click next for more photos)
The way that penguins walk is dictated by their body shape, length of
legs etc.. This is really optimised for swimming rather than walking.
It's more of a making the best of a bad job than anything deliberate.
Oh, and FFS it's GAIT not gate.
| 
del.icio.us
Digg
Google bookmark
reddit
Simpy
StumbleUpon
Furl
Newsvine
Technorati
Tailrank