Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!fas-news.harvard.edu!newspump.wustl.edu!bcm!cs.utexas.edu!howland.reston.ans.net!news.sprintlink.net!pipex!uunet!hobbes!earth.armory.com!rstevew
From: rstevew@armory.com (Richard Steven Walz)
Subject: Re: Biped robot designs & ideas (Question)
Organization: The Armory
Date: Thu, 19 Jan 1995 17:13:10 GMT
Message-ID: <D2nx5y.79I@armory.com>
References: <3evm76$84n@newsbf02.news.aol.com> <yzhu.790285128@saturn.acs.oakland.edu> <D2K3A7.Azt@armory.com> <3fjlqj$175@life.ai.mit.edu>
Sender: news@armory.com (Usenet News)
Nntp-Posting-Host: deepthought.armory.com
Lines: 59

In article <3fjlqj$175@life.ai.mit.edu>,
Peter Dillworth <chunks@peroneus.ai.mit.edu> wrote:
>In article <D2K3A7.Azt@armory.com>, rstevew@armory.com (Richard Steven Walz) writes...
>|> ----------------------------------
>|> I wrote the disparaging remarks. Yes, there are some wonderfully heavy and
>|> high powered tethered bipeds that walk, and if you have a $100K grant you
>|> can build one too! But if you are a hobbyist roboticist or even a small
>|> maker of commerical robots, this technology is out of your reach. The one
>|> legged robot is the cutest but I have not seen it run without a very large
>|> computer and power supply and a heavy multi-conductor tether! The others do
>|> a LOT of falling down. The science shows on the science-Pop network,
>|> Discover channel, make it look good by selective videography, but these
>|> things are not at all ready and able for work quite yet, and that, kids, is
>|> what I meant!
>|> -Steve Walz   rstevew@armory.com
>|> 
>---------------------------------
> 
> This message is far too pessimistic in my opinion.  The state of the art in biped
>robot control is so rudimentary that almost anyone CAN do it, not just the rich and
>powerfull.  Like any technology (take aircraft design as an example), when it is in
>it's conception stages there are much more opportunities for the hobbiest to do very
>good research (the Wright bros. were simple bicycle mechanics, yet they certainly
>performed the most important work in aeronautics in a century).
>
>  I work with biped robots on a full time basis in my lab, and I think I speak from 
>experience, that if a person is handy with electronics, computers, motors, and has
>a desire to see machines walk, that with an expenditure of around $1,000.00 (using
>say model airplane servos for motors, and 6811s for brains), that a simple but
>functional biped robot can be contructed.
>
>  I would highly recommend a new publication called "The Robot Practitioner", to be
>published by FootFalls, ltd.  The first edition will have an artical on just such
>a biped robot, thus prooving my point even further.
>
>happy hacking!
>
>	-Peter Dilworth
>
-------------------------------------------

I wouldn't say that hobbyists can't make some contributions to bipedalism
schemes, that is true. I just say that, compared to treaded and wheeled bots,
that if you ever WANT to do ANYTHING *BUT* bipedal locomotion, that you had
better pick a different transport platform for all the other neat stuff
you'd like on board, power tether or no tether, because I don't see a
feasible bipedal scheme that could carry anything but its own ASS (I
couldn't resist, after "Ice Pirates"). The job is simply nasty enough that
any bipedal *I* have seen was not exactly walking or running up or
downstairs, and none of them are carrying their own main processor or other
manipulators or heavy visual or communicative equipment! And they are
certainly not hiking rough terrain. If you know otherwise, PLEASE show
me!!! Thanks,
-Steve Walz   rstevew@armory.com
P.S, Oh, Peter, can you fix your editor. It's wrapping around the end of
line an making the display of your messages very difficult to read or
respond to. Check with your sysadmin.
-Steve

