🚀 go-pugleaf

In article <1a2ltcINN5e6@koh-sun11.usc.edu>, whu@koh-sun11.usc.edu (Kuo-Ray Whu) writes:
|> 
|> 	Last week I came thru a prof. in control division. He said "
|> 	SOME people do fuzzy control because they don't really know control,"
|> 	Well... I am not control major,so ... How do you think?
|>	How many does the SOME mean?
|>
|> kuoray
|> Dept. of EE
|> USC
|> 
 

Hmmm.  Unfortunate.  :-(

My first reaction is "how much does *he* know about fuzzy
control?"

What is most important in control (as with most other disciplines)
is results.  If using fuzzy control yielded
results than, say, using PID control, does it matter that
you didn't solve differential equations to create your control
system.  (Maybe in school, because testing students would be 
tougher.)  And when I talk about "better results", I'm talking
not only about performance, but development time/cost, 
cost of system, ease of tuning, etc.

I've heard and seen similar comments even in editorial sections
of trade magazines such as the EE Times.  "Fuzzy control is for
engineers who couldn't handle control."  Often times these
comments seem to come from people who don't know fuzzy control.
(And they probably program in assembly language. :-) )

In addition to the consumer electronic sector, fuzzy control is
growing in automotive, process controls, and military/government.
NASA does a lot of work in fuzzy control.  Yeah, and they don't
know control. :-)

You might want to go find Bart Kosko and ask for a second
opinion. :-)   I thought he was at USC.

-- 
Yoke (Yokichi) Tanaka      Togai InfraLogic, Inc.    "fuzzy"
tanaka@til.com             Irvine, California, USA
uunet!til!tanaka           714-975-8522
In case you're wondering ... I am male.
"The more you drive, the less intelligent you are."