Newsgroups: comp.robotics
Path: brunix!sgiblab!swrinde!cs.utexas.edu!howland.reston.ans.net!newsserver.jvnc.net!stevens-tech.edu!vaxc.stevens-tech.edu!jcarnice
From: jcarnice@vaxc.stevens-tech.edu
Subject: Let's talk theory.
Message-ID: <1994Mar25.125259.1@vaxc.stevens-tech.edu>
Lines: 36
Sender: news@dmi.stevens-tech.edu (USENET News System)
Organization: Stevens Institute Of Technology
Date: Fri, 25 Mar 1994 17:52:59 GMT

					<>

	  I'm rather new to NEWS.  I saw this topic and thought it was a place
	for discussing theories and the latest data.  I mostly see requests
	for literature and materials.

	  I've been interested in robotics for a number of years.  I have many
	ideas of how to deal with the mechanical and AI ends of robotics.  But
	I'd like to know what's going on so I don't do any reinventing of the
	wheel and to give some of my own ideas.

	  From what I know about the popular media about the field of robotics,
	it seems that the focus has been on how to program a robot to do lots
	of things like walking, picking up coke cans, and the like.

	  I honestly don't think that preprogramming routines for doing things
	is a good way to go about it.  I have more faith in programming
	robots to have emergent intelligence.  In other words, instead of
	programming a robot to go from point A to point B given terrain type
	C, why not just program it to have the "urge" to go from point A to
	point B and give it the ability to "try" things and keep the ones that
	get it from A to B?

	  I suppose one might say that it would be silly to have to put a
	robot through school before putting it on an assembly line, but there's
	an easy sollution.  School one and copy its knowledge to all others with
	the same architecture.  Simple.

	  This is just a quick overview.  Thoughts?


						-James Vincent Carnicelli II
						 Stevens Institute of Technology
						 Hoboken, New Jersey


