Newsgroups: comp.ai.alife,comp.ai.neural-nets,comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornell!travelers.mail.cornell.edu!news.kei.com!news.mathworks.com!usenet.eel.ufl.edu!news-feed-1.peachnet.edu!gatech!swrinde!cs.utexas.edu!uwm.edu!reuter.cse.ogi.edu!hp-cv!hp-pcd!hpcvsnz!news
From: 5LG3 (Terrill Snyder)
Subject: Re: Gonna make an autonomous, emergent, walking robot (Genghis , style)...how?
Sender: news@hpcvsnz.cv.hp.com (News )
Message-ID: <D54DnK.MHH@hpcvsnz.cv.hp.com>
Date: Wed, 8 Mar 1995 11:38:07 GMT
References: <Ben.19.00276AC0@ins.netins.net> <Pine.ULT.3.91.950228012351.6358D-100000@rac5.wam.umd.edu> <Pine.ULT.3.91.950228022333.9642L-100000-100000@rac5.wam.umd.edu>
Nntp-Posting-Host: seanm.cv.hp.com
Organization: InkJet Business Unit (IJBU)
X-Newsreader: WinVN version 0.82
Lines: 54
Xref: glinda.oz.cs.cmu.edu comp.ai.alife:2742 comp.ai.neural-nets:22672 comp.robotics:18842

In article <Pine.ULT.3.91.950228022333.9642L-100000-100000@rac5.wam.umd.edu>, Keith Wiley <keithw@wam.umd.edu> says:
>
>I've been writing alife stuff for a couple months now and I'm totally
>into emergent behavior more than anything else, and having finished my
>last project (an oversimplified SimLIFE type thing.  Bugs search for
>food, and reproduce evolving search strategies mostly), I have several
>last project (an oversimplified SimLIFE type thing.  Bugs search for
>food, and reproduce evolving search strategies mostly), I have several
>possible plans of courses of action at this point.  I considered a
>flocking algorithm but have lost interest.  I think a robot with
>self-emergent walking behavior would cool.  Here, then, are my two main
>problems:
> 
>1)
>I know very little about neural-networks.  Thanks again to the people who
>have sent my simple, easy to understand descriptions and diagrams.
>However, I still don't know essentially how it's done.  I realize this a
>large problem, as opposed to a small one and I accept full reponsibility
>for figuring this part out.  (Any help is welcome at any time folks).
> 
>2)
>This is where all of you come in guys.  I need to know the approximate
>strategies used in programming Colin Angle's Genghis, NASA's Attila
>(direct descendent of Genghis), Randall Beer's artificial cockroach
>aptly named Periplaneta comptarix, or any other of the several walking
>(direct descendent of Genghis), Randall Beer's artificial cockroach
>aptly named Periplaneta comptarix, or any other of the several walking
>robots and walking simulations out there.  I've been staring pieces of
>paper trying to figure out how to hook up input and output nodes for the
>individual legs with mostly confounding confusion.
> 
>Any and all help is much appreciated, thank you.  I hope to eventually
>bring this program up to the level of a full-scale Mars exploration with
>physical environmental variables and multiple animats working
>collectively to accomplish tasks I give them.
> 
>. . .. ... ..... ........ ............. .....................
>. .. ... ..... ....... ........... ............. .................
>. .. .... ........ ................ ................................
>/*Keith Wiley*/                          *
>#define Electrogenetic Engineer           *   * *     *   *      *
>Str255 residence == Univ. MD;           ***    **   * *    **     *
>contact == email(keithw@wam.umd.edu);          *     **   **    ***
> 
>(incidently, I'll tip my hat (sorry no $.  I'm a college student and
>you'll all get this anyway) to whoever tells me what the three lines of
>periods at the top of my signature/tagline mean)  mostly because I'm
>curious how many people can figure it out.

I think it is too easy of a sequence to stump many people with.

The same sequence can also be generated as a relationship between unity
and a polynomial.  Do you know how?

