Newsgroups: comp.robotics
Path: brunix!news.Brown.EDU!noc.near.net!das-news.harvard.edu!cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!netnews.srv.cs.cmu.edu!gerry
From: gerry@cmu.edu (Gerry Roston)
Subject: Re: "That's not robotics; that's toy building"
In-Reply-To: pdunster@wampyr.cc.uow.edu.au's message of 10 Aug 1993 15:42:25 +1000
Message-ID: <GERRY.93Aug10100832@onion.cmu.edu>
Sender: news@cs.cmu.edu (Usenet News System)
Nntp-Posting-Host: onion.frc.ri.cmu.edu
Reply-To: gerry@cmu.edu (Gerry Roston)
Organization: Field Robotics Center, CMU
References: <CBHzGn.DI6@cs.uiuc.edu> <247cg1$9e4@wampyr.cc.uow.edu.au>
Date: Tue, 10 Aug 1993 14:08:32 GMT
Lines: 57

As a practioner in the field of robots, let me make the following
comments/observations:

1 - Robotics, as an engineering enterprise, can not be classified
    as a "junior branch" of any of the existing fields of engineering,
    despite the attempts of many schools to do so. (Shameless plug 
    alert:)  CMU has wisely created the Robotics Institute, the 
    faculty of which all (?) have cross-postings to other departments,
    including, Mech E, EE, Comp Sci, etc. 

2 - What is a robot is a difficult question to answer. Steve Shaffer
    gives an amusing introductory talk in which he presents the 
    following: He shows (a sketch) of a can opener, and get universal
    accord that it is not a robot. He then shows a progression of
    devices that wind up with the Shaffer slice-o-matic, that 
    does everything in the kitchen that you can imagine, autonomously.
    Everyone agress that this is a robot. The question, though, is
    at which stage did robothood appear?

Defining what is meant by robotics is itself a challenge. My
definition, which is certainly no better than any of the others goes
something like this: A robot is a device that uses sensors to
understand the world it is in and actuators to act on the world using
the information it gathered from its sensors. The fact that this
definition applies to many (most?) biological life forms, i.e.,
animals, does not bother me at all.  (Read Hans Moravec's book... he
blurs the line to an even greater extent.)  However, it has been
pointed out that my definition also describes a toaster, which is
certainly not (?) a robot.

Now personally, I feel that most hobbiest who claim to be building
robots aren't. They are making toys, clever toys, expensive toys, but
toys none-the-less. I guess that the reason I feel this way is that a
crucial element of a robot is the need for it to have a purpose. The
robot we are currently building is an Earth analog for a robot that
could explore the lunar surface for at least one complete lunar day.
This is a defining purpose. To simply build a mechanical device that
wanders around a room without hitting the walls, well maybe several
years ago that was a robot, the the field has moved past that point.

Since like most of us, I am thinking as I write, I think the last
sentence is a key element to the definition of a robot.  I believe the
definition may be "time dependent". For instance, if R Daniel were to
show up in the lab tomorrow, I would be hard pressed to refer to our
latest efforts as robots when presented with R Daniel.

--
Gerry Roston (gerry@cmu.edu)    | Because experience witnesseth that eccle-
Field Robotics Center,          | siastical establishments, instead of main-
Carnegie Mellon University      | taining the purity and efficacy of Religion,
Pittsburgh, PA, 15213           | have had a contrary operation.  During almost
(412) 268-3856                  | fifteen centuries has the legal establishment
                                | of Christianity been on trial.  What have been
The opinions expressed are mine | its fruits?  More or less in all places, pride
and do not reflect the official | and indolence in the Clergy, ignorance and
position of CMU, FRC, RedZone,  | servility in the laity, in both, superstition,
or any other organization.      | bigotry and persecution.  James Madison
