Subject:
|
Re: Robot intelligence
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Mon, 15 Mar 1999 21:32:49 GMT
|
Original-From:
|
B van Dam <BVANDAM@spamlessWESTBRABANT.NET>
|
Viewed:
|
1043 times
|
| |
|
|
Hi Mario,
I'm trying to design a robot pet (using the Cybermaster and neural brain AI
software in VB). The problem I have encountered is that apart from the
basics (the desire for 'food') it is difficult to find suitable targets for
it's behaviour. How for example do I get it to explore it's world, how do
you create a desire to mate etc, etc. Is this described in the books you're
referring to? And are these books on the internet somewhere?
CU Bert
Visit my site on Lego, Robot's and Visual Basic at
http://home.concepts.nl/~bvandam/index.html
-----Oorspronkelijk bericht-----
Van: Mario Ferrari <mario.ferrari@edis.it>
Aan: lego-robotics@crynwr.com <lego-robotics@crynwr.com>
Datum: maandag 15 maart 1999 16:17
Onderwerp: Robot intelligence
> In the book "How to design and build your own custom robot" by David L.
> Heiserman, Tab Books, I found a classification of robots according to their
> "intelligence" (pages from 11 to 17). I wonder if this is just the author's
> thought or if there's a sort general consensus on this theme. I personally
> find this categories meaningful and useful.
>
> I try to summarise some of the ideas:
>
> PARABOT – it's a machine that can behave a fashion similar to that of a • robot,
> but is not a robot in the truest sense, because it lacks an element of
> autonomy or self-determination. Heiserman puts in this class all the • machines
> that are directly or indirectly controlled by a human operator. Most of the
> "industrial robots" fall here: they can perform complex tasks but need • human
> intervention at some point to switch from a task to another.
>
> ROBOT – If a machine is something more then a parabot, it's a robot by
> default. Key words are "autonomy" and "adaptability".
>
> Alpha-Class Intelligence – A machine that has absolutely no capacity for
> remembering past experiences; it exists only in a very narrow time frame of
> the moment.
>
> Beta-Class Intelligence – A machine with the ability to remember responses
> from past encounters with the environment, and then call up those responses
> when similar conditions arise later. A Beta-Class machine actually works on • an
> Alpha-Class level through the early part of its life, but as its hierarchy • of
> experience grows, it exhibit smaller amounts of purely random behaviour in
> favour of "habits" it has learned.
>
> Gamma-Class Intelligence – A machine that starts its life as an Alpha and • soon
> begins exhibiting the memory-related programming of a Beta. Then, as the
> hierarchy of remembered responses grows and becomes refined with further
> experience, begins generalising that knowledge to situations not yet
> encountered on a first-hand basis.
>
> There aren't examples of Gamma-Class robots in the book, but the author • writes
> he has published programs for engendering that sort of behaviour in the • books
> "How to build your own self-programmed robot" and "Robot intelligence…with
> experiments".
>
> Mario
>
> ----------------------------------------------------
> Mario Ferrari
> http://www.geocities.com/CapeCanaveral/Galaxy/9449/
> e-mail: mario.ferrari@edis.it
> ----------------------------------------------------
> --
> Did you check the web site first?: http://www.crynwr.com/lego-robotics
--
Did you check the web site first?: http://www.crynwr.com/lego-robotics
|
|
Message has 1 Reply:
 | | Re: Robot intelligence
|
| (...) It's all a matter of framing it in the correct reward/punishment context. E.g., if you want it to learn its environment, punish it for not running into things, reward it for actively avoiding obstacles (which opens the question: How're you (...) (26 years ago, 15-Mar-99, to lugnet.robotics)
|
5 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
