Newsgroups: comp.robotics
Path: brunix!uunet!decwrl!netcomsv!mork!nagle
From: nagle@netcom.com (John Nagle)
Subject: Re: Tentacle robots.
Message-ID: <tg#mq-k.nagle@netcom.com>
Date: Mon, 13 Jul 92 05:56:14 GMT
Organization: Netcom - Online Communication Services  (408 241-9760 guest) 
References: <1992Jul11.173902.97351@cs.cmu.edu> <1992Jul13.001250.7008@monu1.cc.monash.edu.au>
Lines: 28

jak@monu1.cc.monash.edu.au (mr j grimm) writes:

>This is probably a silly question, but has anyone worked on tentacle
>robots. I envisage a sort of tube of material that has sections that
>compress (or expand) when an electrical impulse is applied. Is there some
>sort of electrically sensitive material that could do the job ?

       Things that look like tentacles have been built as industrial
spray-painting robots.  The mechanism, though, is conventional, with
a multi-segmented arm controlled by cables driven by actuators in the base.

       Various technologies for making materials that change shape on
command have been tried, but so far, nothing really works very well.
There's "NiTiNol", the "memory metal" alloy, but power output is low.
Compressed air in tubes inside webbing produces a muscle-like effect,
but is far less powerful than an air cylinder.  There are fluids that
change cohesion when electric or magnetic fields are applied, but these
seem best suited to clutches.  Piezoelectricity can produce considerable
force but not much travel, although there is a Japanese rotating pizeoelectric
motor prototype.  

       Electric motors, especially modern brushless servomotors with 
good controllers, seem to be the actuator of choice for now for machines
at typical robot scales.  (Micromachines are a whole different story.
Piezoelectricity and electrostatics are very useful there, while
magnetics isn't.)

					John Nagle
