Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.ai.fuzzy    |    Fuzzy logic... all warm and fuzzy-like    |    1,275 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 427 of 1,275    |
|    Walter Banks to Oliver Bandel    |
|    Re: Fuzzy Logic vs. Fuzzy Numbers/values    |
|    21 Mar 05 12:26:18    |
      From: walter@bytecraft.com              Oliver Bandel wrote:              > >>Fuzzy numbers may be a nice thing and well suited       > >>for some things...       > >>       > >>...but can fuzzy logic gain an advantage out of       > >>the fuzzy numbers?              Linguistic variables used in fuzzy logic can be viewed as a normalized form of       the problems space. In that sense they often have a computational advantage       over crisp calculations.              Linguistic variables are normalized between 0 and 1 and the precision of the       calculations depends on the number of bits used to represent a linguistic       variable.              In practical applications linguistic numbers are used to the very end of       control algorithms where the linguistic variable is translated back into a       crisp control. This could be a simple on/off switch or a power setting between       0 and 1 or or 0 and 100. the       final point being that once the control algorithm is complete the translation       back to the application specific numbers is usually very easy.                     > .. so it's not a practical issue to use fuzzy numbers       > for calculating/describing/explaining fuzzy controllers?       > They are better calculated with crisp numbers?       >       > Or is it only the lack of computational support by (missing)       > libraries that yields to "going back to what we know better"?!              It is generally less computationally intense to use Linguistic variables than       crisp in most control applications. The reason is once a crisp number is       normalized to a Linguistic variables in a well designed system the computation       requirements depend on       the requirements of the result usually with a smaller dynamic range that the       original crisp numbers.              There are some fuzzy logic examples including control on Byte Craft's Web site              http://www.bytecraft.com/publishing.html              Walter..              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca