Search
 
 

Display results as :
 


Rechercher Advanced Search

Latest topics
» Cherry Blossom Technology Launch!!!
Sat Mar 25, 2017 10:42 am by BlueSodaX

» Will aiko V2+ can/will be able to have sex ?
Sat Jul 25, 2015 6:08 am by Lordoomer

» Is this project still going?
Tue Jul 21, 2015 9:54 pm by Superdog1138

» Aiko New Hand V2 from 3d printer
Thu Apr 02, 2015 7:44 am by Kthomp06

» Tip: Try to create a start-up
Fri Sep 19, 2014 6:03 pm by Kazama_Tenkai

» Hello Everyone!
Mon Feb 03, 2014 6:01 pm by BlueSodaX

» My best regards to Mr. Trungs herculean endeavor
Sun Jan 12, 2014 2:09 am by Discoman

» how to make robots that are more accepted by people-through war.
Mon Dec 16, 2013 11:22 pm by Boetharch

» Who want to see Sumomo From Chobits made?
Sat Sep 21, 2013 7:30 pm by minoru

October 2017
MonTueWedThuFriSatSun
      1
2345678
9101112131415
16171819202122
23242526272829
3031     

Calendar Calendar


The Role of Fuzziness in Artifical Intelligence

View previous topic View next topic Go down

The Role of Fuzziness in Artifical Intelligence

Post  Gilda on Sat May 07, 2011 10:36 pm

Mark Stosberg
12/16/96
Minds and Machines found in http://mark.stosberg.com/Tech/fuzzy/role_in_ai.html

The Role of Fuzziness in Artifical Intelligence



Fuzzy logic is perhaps the most promising advancement to come along in the Artifical Intelligence community in recent history. What exactly lies behind the term "fuzzy", what can fuzziness bring to the advancement of AI and what would fuzzy AI mean for the future?
Fuzzy logic in it's simplest terms expands the dicotomy of true or not true to include a range of answers in between. The usual example is say instead of being black or white, fuzziness allows for shades of gray. Since fuzzy logic allows this extra bandwidth in fuzzy answers, fuzzy rules used in programming can cover a much broader area. A fuzzy rule such as "When it rains, you get wet"*** can cover a lot of ground. It would be able to several instantions of itself such as "when it rains a lot, you get wet a lot" or "when it rains a little, you get wet a little".

Rules like this are beautiful because they are human rules. They are a much better model of how we think. It is not often that questions that arise in life have bivalent answers. There are a few that do such as "Are you married?". Other questions such as "Do you like your job?", would tend to elicit a range of a response falling somewhere between yes and no.

How exactly is a fuzzy rule able to cover so much ground? By the use of a patches. A fuzzy rule will define a fuzzy patch. Say for example that you would like to use fuzzy logic to control an air conditioner (Kosco's example). You could define a fuzzy set for the temparature range as COLD, COOL, JUST RIGHT, WARM and HOT. A system could be composed of a few sloppy rules with wide patches, or many precise rules with narrow patches. Perhaps the air conditioner system is representative of other real systems. That is, an optimal solution involves some wide sloppy rules, and some precise ones.

The fuzzy rules that would go with the air conditioner system would be:

Rule 1: If the temperature is cold, them motor speed stops.
Rule 2: If the temperature is cool, the motor speed slows.
Rule 3: If the temperature is just right, the motor speed is medium.
Rule 4: If the tempertaure is warm, the motor speed is fast.
Rule 5: If the temperature is hot, the motor speed blasts.

This fuzzy system works well because the patches will cover lines that correspond to relations between temperature and motor speed if they are non-linear and squiggle. In contrast a similar bivalent system might be built of many specific rules such as if temparture is 60 degrees than the motor speed is 50. Patches developed from rules like this would only be points, and the system developed from it would only define a collection of points-- not a terrific model. Therein lies a greatness of fuzziness.

Another wonderful aspect of fuzziness is that it does not disclude the old bivalent system of logic. The fuzzy spectrum of greys, completely true and completely false simply fit in as black and white. That the old system fits into the new fuzzy system gives me the feeling the old way was on the right track, and fuzzy logic has added a robustness and ingenious efficiency on top of the old system. Also it allows to move forward slowy from the old system rather than taking a radical step away from it.

If fuzziness is so wonderful, what can it really contribute to the development of AI? Let's first look at some things that have already been developed with fuzzy logic. First of all, there are the microwaves in the dorms here that can cook perfect popcorn. There are air conditions with similar rules sets as described above. This system prevents overshoot-undershoot temperature oscillation and consumes less power. There is an auto transmission that uses fuzzy logic to select gear ratio based on engine load, driving style, and road conditions. There is fuzzy factory control software that schedules taks and assembly line strategies. There is even a fuzzy golf diagnotic system that selects golf clubs based on golfer's physique and swing. The list is lengthy and amazing. From toasters to train systems, fuzzy logic is making machines "smarter".

To make apparent the contribution that fuzzy logic can make to AI (here I am thinking of human AI), let's look at some of the products that display intelligence in more humanlike tasks. Sony has developed a fuzzy based palm top computer that can recognize handwritten Kanji characters. Epson has made a translator that recognizes and translates words in a pencil sized unit. Most interesting was a washing machine that adjusts its washing strategy based on sensed dirt level, fabric type, load size and water level and used a neural network to tune the rules to a user's tastes.

This kind of system is an example of adaptive fuzzy logic. With the help of a neural net, it can learn from the data it has collected and adjust its rules. This kind of set up has tremendous possibilities.

Fuzzy Thinking : The New Science of Fuzzy Logic was the primary source for this report.
avatar
Gilda

Posts : 127
Reputation : 0
Join date : 2010-10-22
Age : 32
Location : Brazil

View user profile

Back to top Go down

Re: The Role of Fuzziness in Artifical Intelligence

Post  Aiko on Mon May 16, 2011 4:49 pm

Hmm..

As for as I know all robot kits that you build are base on fuzzy logic, some are simple and other are more complex.
A good example is maze robot.

As for Aiko BRAINS software it has many fuzzy logic.
Here is example of checking weather.

The BRAINS check the weather locally first. then process it and give information back to the people.
It check for temp, pressure change, cloudy, sunny, windy,foggy, pressure change, humnity...etc.
it has to predict what those informations above and what will be the outcome....

if (WeatherDetails == "outside")
{
string tempstring = "";

if (WeatherOutsideWind > 45)
{
tempstring = "Please dress warmly, It's very windy outside. Be careful when you are going outside.";

}
else if (WeatherOutsideWind > 29)
{
tempstring = "Please dress warmly, It's windy outside. Have a nice day, and be careful.";
}

else if (WeatherOutsideTemp < 0)
{
tempstring = "It's very cold outside. Make sure you have extra clothing. Have a nice day, and be careful.";
}
else if (WeatherOutsideTemp < 10)
{
tempstring = "It's a bit cool outside. Have a nice day.";
}

else if (WeatherOutsideCloud == "FogDay")
{
tempstring = "It's foggy outside. Please watch out for car.";
}
else if (WeatherOutsideCloud == "SnowDay")
{
tempstring = "It's Snowing outside. You may want to dress warmly. Please be careful going outside.";
}
else if (WeatherOutsideCloud == "RainDay")
{
tempstring = "It's Raining outside. You may want to bring an umberlla with you. Please be careful going outside.";
}

else if (WeatherOutsideCloud == "CloudDay")
{
tempstring = "It's Coudy outside. It may rains so take care.";
}
else if
{
tempstring = "It's a very nice day, please enjoy going outside.";
}
.
.
.


etc..

There are about 800 lines of above codes just for the weather logic.

I hope that make sense.... it's simple but it does the job.
avatar
Aiko
Admin

Posts : 335
Reputation : 2
Join date : 2008-08-22

View user profile http://projectaiko.com

Back to top Go down

Re: The Role of Fuzziness in Artifical Intelligence

Post  Gilda on Tue May 17, 2011 11:53 pm

Trung, it is impressive. I will allways say that you are brilliant.
I'd love to help you more and I hope one day I will be able to, but now ... well, it is not so possible, so I go on in my researches in internet.

May I ask you one thing?
About weather predicton she uses a kind of internal barometer or uses internet?

Fuzzy possibilities are amazing. Actually I am reading some papers about this subject.
Did you receive one paper about fuzzy logic and simulation of feelings?
Aiko V2 is able to cry, so mabe she can "feel" conditioned feelings, using some key words .
Example:
Maybe she can feel some feelings conditioned. This could be possible if some words were set in order of priority. Example: To Love> like very much> like> like a little> dislike> detest.
So if a person say any of the kay words she can "feel" express in a better way what a person would expect from a given sentence. Touching or talking the person in a humanlike way.
I imagine that Fuzzy could be used also to determine the order of "friendship" with a family member: Master( or lover, or boss, or any other name people want to give to the first class owner) >friend ( maybe there is more than one person)> fellow( more then one person)> unknown person> enemie ( if a person pokes her she can express "afraid", cry, ask for protection - like in David - from AI- or if she is a security agent she can be agressive)... and so on...

Are these ideas good?
What do you think, is it possible to do?
avatar
Gilda

Posts : 127
Reputation : 0
Join date : 2010-10-22
Age : 32
Location : Brazil

View user profile

Back to top Go down

Re: The Role of Fuzziness in Artifical Intelligence

Post  Aiko on Wed May 18, 2011 12:20 am

HI

I have internal temp sensor, and huminity sensor, and pressure change sensors.
However, if internet access is available the Brains software automatically gets the information from government statelitte for weather change around the world.
from temperature, to the height of the cloud, to pressure from 3000meter above or sea level.etc... to precipitation....or your 5 days weather forcast.
Any city around the world instantly.
Kind of having my own personal weather lady around me.


AS for fuzzy logic for emotion.
I can never figure out how to do that in computer language.
I have never read any paper on fuzzy logic or emotion simulation paper for AI.
I have been busy reading medical books for the last few years, don't really have the time to read anything else.

Maybe when I have the funding or win a lottery ticket, and don't worry about money for food.
I will sit down and spent 6 months writing codes for fuzzy logic for emotion simulation for BRAINS software.


But I see where u are going with Aiko's crying ability.
There is a lot of option for that.....
But it's a working progress...
Just keep giving me idea... and help to make Aiko more lively.
Good job, keep up the reserach for me.
avatar
Aiko
Admin

Posts : 335
Reputation : 2
Join date : 2008-08-22

View user profile http://projectaiko.com

Back to top Go down

Re: The Role of Fuzziness in Artifical Intelligence

Post  Discoman on Wed May 18, 2011 11:13 pm

as far as emotions go, I think David Levy is running some research into this area.
didn't he write a book on human robot interactions?
or am i getting a bunch of stuff mixed up?

can't really verify that, as my router isn't wanting to let most websites through. luckily this one made it Very Happy
avatar
Discoman
Admin

Posts : 212
Reputation : 0
Join date : 2009-07-01
Age : 24
Location : USA

View user profile http://blogofthehawk.com

Back to top Go down

Re: The Role of Fuzziness in Artifical Intelligence

Post  Artilects rule! on Wed Oct 26, 2011 1:09 pm

The fuzzy rules that would go with the air conditioner system would be:

Rule 1: If the temperature is cold, the motor speed stops.
Rule 2: If the temperature is cool, the motor speed slows.
Rule 3: If the temperature is just right, the motor speed is medium.
Rule 4: If the temperature is warm, the motor speed is fast.
Rule 5: If the temperature is hot, the motor speed blasts.
"I don't know that word."
Ok, I get that new settings are within the general instructions, but how would it know the outcome?
For instance "just right," what does that mean to the machine?
The user must input this knowledge.

"What does blast mean?"
Blast is just a word to it with no real meaning.
A better command would be "use extra motor speed," but again, the user must input what is extra speed, 96 percent, 98 percent, or 100 percent (100 percent eventually breaks the machine).

A better term for medium is "half power," like in microwave instructions.

Of course these user set terms would be permanent, so in that way it could know the meaning, such as a "Smart Phone".

Artilects rule!

Posts : 202
Reputation : 0
Join date : 2009-02-06
Location : United States

View user profile

Back to top Go down

Re: The Role of Fuzziness in Artifical Intelligence

Post  Sponsored content


Sponsored content


Back to top Go down

View previous topic View next topic Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum