Talk:Fuzzy control system

From Wikipedia, the free encyclopedia

This looks like a great article, but I'm not convinced fuzzy logic should be redirected here. Perhaps the article should be split into two parts - one about the logic (which should be in an article titled "fuzzy logic") and one about the control systems (which, after all, are an *appplication* of fuzzy logic). --Robert Merkel

I think they should be merged back. I didn't even know about this article until just now. Yes, fuzzy control is an application of fuzzy logic, but there's little here that's specific to control. I mean, the whole notion of fuzzy sets and fuzzy rules can be applied to virtually anything. They are fuzzy logic. "Fuzzy logic" is a bit of a misnomer since it has more to do with the set theory than actual "logic". - furrykef (Talk at me) 05:26, 3 Mar 2005 (UTC)

I disagree. There is a lot here that is specific to control. For example, fuzzy logic itself does not necessarily need to deal with the processes of fuzzification and defuzzification (although these are important issues in most applications), and the block diagrams here are a good example of content that is specific to fuzzy control.

On an unrelated note, it would be cool to have a section explaining the differences between fuzzy control and other approaches to control systems. Perhaps even a note about that in the intro. CyborgTosser (Only half the battle) 19:39, 6 May 2005 (UTC)

---

ASCII art to proper art, please :) Martin

• I copied over most of th images from th original article (which are all explicitly public domain), but th last one didn't seem to fit entirely with its ASCII counterpart, and one had already been re-created by a Wikipedian. Boffy b 07:03, 16 Nov 2003 (UTC)

[edit] PID vs Fuzzy

In the section "Fuzzy Sets" the author has made a reference to the PID controller and has put forward the obvious question: why do we need fuzzy controller. I think the answer is far from being scientific or true. If one has a low resolution analog digital converter the input of the PID controller will be within this resolution. If you can measure the temperature of a break with a resolution of 1/4 then you would have a low resolution input to your PID controller. Again if your actuator can move only within a space of four steps then your output will be a number between 1 and 4. Suggesting that fuzzy control's advantage is that it is suitable for cheap and low resolution systems is total nonsense. The suggestion that "a mathematical model of the control process may not exist, or may be too "expensive" in terms of computer processing power and memory" sounds childish. What expense would there be in calculating a z-transformation of a PID controller and implementing a digital controller which will have to keep the 4th or 5th past values in its memory??? --Zama Zalotta 10:01, 6 February 2007 (UTC)

[edit] Linguistic Variable

Well-written article. What should be underscored is that there are two constructs which play pivotal roles in fuzzy control and, more generally, in most applications of fuzzy logic. They are (a) the concept of a linguistic variable; and (b) the calculus of fuzzy-if-then rules. These constructs were introduced in my 1973 paper, "Outline of a new approach to the analysis of complex systems and decision processes," IEEE Transactions on Systems, Man and Cybernetics SMC-3, 28-44, 1973, and my 1974 paper, "On the analysis of large scale systems," Systems Approaches and Environment Problems, H. Gottinger (ed.), 23-37. Gottingen: Vandenhoeck and Ruprecht, 1974. It is of historical interest to note that the concept of linguistic variable was greeted with skepticism and derision. On hearing my first presentation in France in 1972, Rudy Kalman (Kalman filtering) said, "I would like to comment briefly on Professor Zadeh’s presentation. His proposals could be severely, ferociously, even brutally criticized from a technical point of view. This would be out of place here. But a blunt question remains: Is Professor Zadeh presenting important ideas or is he indulging in wishful thinking?" Fuzzy logic and fuzzy control have survived such comments, including highly critical comments by the control systems establishment.

A metric of the impact of fuzzy logic is the number of papers and books with "fuzzy" in title. The following information was compiled by Camille Wanat, Engineering Librarian, UC Berkeley, March 13, 2007.

INSPEC Database - "fuzzy" in the title

1970-1979: 569

1980-1989: 2,403

1990-1999: 23,214

2000-present: 24,910

Total: 51,096

MathSciNet Database - "fuzzy" in the title

1970-1979: 443

1980-1989: 2,465

1990-1999: 5,487

2000-present: 6,217

Total: 14,612

--Lotfi A. Zadeh 01:23, 15 March 2007 (UTC)