Society of Robots - Robot Forum

General Misc => Misc => Topic started by: Admin on April 29, 2007, 02:08:57 PM

Title: SoR's new definition for a robot
Post by: Admin on April 29, 2007, 02:08:57 PM
We talked a lot about a what the definition for a robot is in a previous post here:
http://www.societyofrobots.com/robotforum/index.php?topic=540.0

Basically we came to the conclusion that there is no good definition . . .

But after thinking about it for awhile, I got a very GOOD definition and I want feedback on it.

SoR's definition of a robot:
An anthropomorphic electromechanical machine that can interact with the environment.

The key word here is anthropomorphic (http://en.wikipedia.org/wiki/Anthropomorphism). Why? Because I think that if people cannot connect with the machine as a 'living' entity then it just doesnt 'look' like a robot.

For example:
A computer is not a robot, but a computer with wheels is.
A washing machine isnt a robot, unless it had arms to load your clothes.
Your car is not a robot, unless it drove itself.
An automated donut making machine is not a robot, but a biped serving you donuts is.

Note that it doesnt need to be automous, or actually even sense the environment. It just needs to interact with the environment. For example,
A remote control battle bot is a robot, despite not having any brains.
A puppet is not a robot because it is not an electromechanical machine.

The point I am trying to make is, the machine must act like a 'living' entity for it to be considered a robot.

Thoughts anyone? Flaws in my logic? ???
Title: Re: SoR's new definition for a robot
Post by: ed1380 on April 29, 2007, 02:32:01 PM
sounds good to me, except for the battle bots, cause they're more like remote controlled cars. the only reason rc cars are called cars not robots is cause of looks, reason battle bots are called bots is cause they don't look like cars
Title: Re: SoR's new definition for a robot
Post by: Brandon121233 on April 29, 2007, 03:33:11 PM
I still think that it is necessary to have a separate definition for autonomous and human operated robots, they seem way too far apart to be defined as the same thing, and in trying to find a definition that does encompass both of them, I think that allows some things in that are definitely not robots. Just my thoughts. 
Title: Re: SoR's new definition for a robot
Post by: myblack60impala on April 30, 2007, 01:10:36 AM
I agree with your opinions on what a robot is. So I have a question, do you think this picture is a robot?

(http://www.winston.robomo.com/obj26geo25pg1p4.png)

I built this as a cheap wall follower, it uses no brains, its just a battery and 2 parallel wired motors, and a bearing caster. It works by using offset power delivery to hug the right wall of a maze, then when presented a left turn it uses the front wheel to make the turn. If its not against a wall, it just does circles.
Title: Re: SoR's new definition for a robot
Post by: Admin on April 30, 2007, 05:01:43 AM
Quote
the only reason rc cars are called cars not robots is cause of looks, reason battle bots are called bots is cause they don't look like cars
hmmmm good point . . .

Quote
in trying to find a definition that does encompass both of them, I think that allows some things in that are definitely not robots.
examples?

Quote
So I have a question, do you think this picture is a robot?
yes, this would qualify in my new definition
Title: Re: SoR's new definition for a robot
Post by: Kohanbash on April 30, 2007, 11:06:40 AM
What about a typewriter? It has human characteristics, its an electromechanical machine, and it interacts with its environment.
Title: Re: SoR's new definition for a robot
Post by: ed1380 on April 30, 2007, 04:08:47 PM
some are electromechanical, but I think it has to be anonymous, or something to that extent. like if it writes your report for you ;D
Title: Re: SoR's new definition for a robot
Post by: Brandon121233 on April 30, 2007, 04:11:21 PM
Your furnace or A/C, it has electromechanical parts, interacts with its environment (when its too cold it turns on and vice versa), and many can argue that it looks like a robot with all those arms.
Title: Re: SoR's new definition for a robot
Post by: Admin on April 30, 2007, 04:22:17 PM
Ok, so i guess a major flaw in my new definition is that the term 'anthropomorphic' is very subjective.

To me, both a typewriter and the AC arent anthropomorphic . . . but I still think Ive made a good point why people consider battle bots a robot but washing machines not . . .

Brandon, your AC has arms?!
Title: Re: SoR's new definition for a robot
Post by: Brandon121233 on April 30, 2007, 04:30:04 PM
LOL, no I think I'm trying to stretch it too far (I meant all the ducts comming out of it kind look like arms), I think your definition is very good, but I still think the term "robot" is extremely loose and open to be defined by and person differently. There will always be disputes as to what is and is not a robot.
Title: Re: SoR's new definition for a robot
Post by: gamefreak on April 30, 2007, 07:23:55 PM
do you count a human controlled device that can move around/interact withen the environment a robot?
not autonomous in any way.
Title: Re: SoR's new definition for a robot
Post by: mpthompson on May 02, 2007, 12:53:17 AM
I was at a presentation last week regarding the history of robotics and the basic definition the speaker gave for a robot was "an inanimate object that is 'surprisingly' animate".  Such a definition is designed to last the test of time.  For instance, today we wouldn't think of washing machines, escalators and automatic opening doors at the supermarket as robots, but to someone in the 19th century they very well would be seen as robots, if not magical.  Likewise, I suspect that to our grandchildren, our current concepts of robots may ultimately seem naive and pedestrian.

-Mike