Fuzzy Logic Operating Systems - Embedded

This is a discussion on Fuzzy Logic Operating Systems - Embedded ; Hi everybody! I am currently a research associate at the Vienna University of Technology & responsible for the creation of new projects. Currently, I have a new subject that involves fuzzy control and fuzzy logic operating systems. What I want ...

+ Reply to Thread
Results 1 to 15 of 15

Thread: Fuzzy Logic Operating Systems

  1. Fuzzy Logic Operating Systems

    Hi everybody!

    I am currently a research associate at the Vienna University of Technology &
    responsible for the creation of new projects. Currently, I have a new
    subject that involves fuzzy control and fuzzy logic operating systems.

    What I want to discuss here is the reasons for using a fuzzy logic operating
    system that makes some decisions using the theory of fuzzy logics.

    =================
    To incorporate the fuzzy idea into the operating system is not a devious
    idea to me, as complex systems do not have necessarily crisp arbitration.
    With *complex systems*, I do not only mean expert systems or other types of
    AI, but relatively "simple" systems like OS. Most often, (OS) designs do not
    incorporate uncertainties at all, and the computer languages used to
    implement OSs are using crisp logic. For instance, a C expression will
    either be matched (true) or not (false). If this is the way OS have been
    designed, the question is if this the (best) right way to do so.

    The idea is to use fuzzy logic in OS design is not new. On the one hand,
    some FLOS products were already on the market (FuzzySoft AG), but the
    products were not accepted by customers (see
    http://www.dbai.tuwien.ac.at/marchiv...il95/0734.html ). On the
    other hand, some of these OS's could even make it and were used in fuzzy
    control systems (Bosch WET2820GB (BCH-WET2820GB) Built-In Washer Dryer,
    WFE2021GB, etc.). And there exists research to use FLOS even in hard
    real-time systems [1]!

    The reason fuzzy logic is used are:
    1. Complex systems cannot be pinned down to a set of formulas easily.
    2. It is very difficult to write down the knowledge of an expert in crisp,
    Boolean logic. The propositional logic cannot handle more complicated cases
    like believe revision, which may be necessary if several experts are asked.
    3. Last but not least, humans think fuzzy. We express ourselves using fuzzy
    terms, called words. The effort to simplify the reality by mapping it to
    Boolean logic results sometimes in problems that are later found and their
    correction is expensive. And crisp complex systems do not behave optimal
    when confronted with the reality.

    OS are responsible to provide the limited resources of the system to the
    processes. For real-time OS's, they have not only to assign the resources
    but they have to guarantee a response time for each request as well. To
    avoid problems like deadlocks, limits must be defined that determine when a
    response time was too late, or whether there were other occurances that are
    not allowed. These criteria can be defined with crisp logic, "But most
    real-world application's knowledge is not so hard. The boundaries are more
    of a softer kind. An insignificant violation of a constraint may cause a
    crisp constraint satisfaction system to reject a schedule that could have
    been the best )beside the insignificant violation) to find. And that is the
    point where fuzzy theory is involved. Soft constraints are modelled by
    flexible fuzzy constraints" [1].

    Another point is the constraint satisfaction. There are cases where not all
    constraints have to be satisfied. "Release dates and due dates of jobs and
    operations are obvious candidates for fuzzification. An operation may start
    a little earlier or later and a small due date violation may be acceptable
    either."

    Also, at a higher level, a fuzzyfication may make sense. For instance, if I
    make a search for all files containing the word "radnom" in the file name,
    the file system could ask whether this should be "random". As a person who
    can understand five languages, it is also something I would appreciate alot
    if a word or subject searched is also searched in other languages the user
    knows. Also, a verb may have different declinations that are also
    interesting in a search. Another point is when searching all files having
    .... that are greater than 10 MB, say. Are here files are only slightly
    smaller than this minimum (256 bytes less than 10 MB) really uninteresting
    to the user?

    We all know some helps used in sophisticated tools, like MS Office, that
    help alot when typing text. One of these features is the auto correction of
    words. This could be a service supplied by the OS so that all the software
    may make use of it. The price of making it available for the OS designer
    would increase the costs of the OS, but encapsulate this feature to it and
    will make all software better usable & with less code lines. So, the
    software for this OS will become more stable. Oh sorry that I forgot to say
    the link to fuzzyness concerning auto correction this time! But to decide
    which language is to be corrected is a fuzzy concern. If there is a citation
    of another language, the system may find out - as the reader also can find
    out the original and the cited text from a different language - using fuzzy
    logic.


    I wrote these patches down only to start a discussion on this subject. The
    questions that may be discussed are:
    (1) Is it reasonable to fuzzify operating systems?
    (2) Where can such fuzzification procedures give improvements? What are the
    problems in crisp OS'es that would not be a problem for FLOS?
    (3) What FLOS have been used designed since now? What are their outstanding
    properties? What was the benefit of using fuzzy logic in design,
    development, testing? How did these systems react when confronted with real
    problems?
    (4) What about using FLOS in (hard/soft) real-time systems? What are the
    benefits? Where can fuzzy definitely not be used?
    (5) What about the impact on cost of OS development/implementation/testing?
    (6) ...


    ================================================== ===========
    [1] « Fuzzy and Possibilistic Approaches towards Scheduling Applications»,
    Vienna University of Technology, diploma thesis of Thomas Länger,
    1994-04-26.


    --

    Kind regards/Mit freundlichen Grüßen,

    D.I. Ekrem SABAN
    Vienna University of Technology/TU Wien
    Institute for Computer-Aided Automation/Institut für Rechnergestützte
    Automation
    Department for Automation Systems/Abteilung für Automatisierungssysteme
    (+43-1)58 801-183 23

    http://reopen911.org



  2. Re: Fuzzy Logic Operating Systems

    On Mon, 13 Feb 2006 15:41:12 +0100, Ekrem SABAN wrote:

    > We all know some helps used in sophisticated tools, like MS Office

    ^^^^^^^^^^^^^^^^^^^

    Damn. I was with you up to this point, but here I think you misplat
    "bloated PoS".

    --
    mark south: world citizen, net denizen
    echo znexfbhgu2000@lnubb.pb.hx|tr a-z n-za-m
    "Take it? I can't even parse it!" - Kibo, in ARK


  3. Re: Fuzzy Logic Operating Systems

    On Monday 13 February 2006 16:58, Mark South stood up and spoke the
    following words to the masses in /comp.os.linux.misc...:/

    > On Mon, 13 Feb 2006 15:41:12 +0100, Ekrem SABAN wrote:
    >
    >> We all know some helps used in sophisticated tools, like MS Office

    > ^^^^^^^^^^^^^^^^^^^
    >
    > Damn. I was with you up to this point, but here I think you misplat
    > "bloated PoS".


    We could also inform the OP that if he wants a fuzzy operating system,
    he has to turn to Windows. Anything coming from Microsoft is as fuzzy
    as it gets...

    --
    With kind regards,

    *Aragorn*
    (Registered GNU/Linux user #223157)

  4. Re: Fuzzy Logic Operating Systems

    On Mon, 13 Feb 2006 17:49:25 +0000, Aragorn wrote:

    > On Monday 13 February 2006 16:58, Mark South stood up and spoke the
    > following words to the masses in /comp.os.linux.misc...:/
    >
    >> On Mon, 13 Feb 2006 15:41:12 +0100, Ekrem SABAN wrote:
    >>
    >>> We all know some helps used in sophisticated tools, like MS Office

    >> ^^^^^^^^^^^^^^^^^^^
    >>
    >> Damn. I was with you up to this point, but here I think you misplat
    >> "bloated PoS".

    >
    > We could also inform the OP that if he wants a fuzzy operating system,
    > he has to turn to Windows. Anything coming from Microsoft is as fuzzy
    > as it gets...


    Yeah, Linux system don't have fuzz, just lint. .
    --
    mark south: world citizen, net denizen
    echo znexfbhgu2000@lnubb.pb.hx|tr a-z n-za-m
    "Take it? I can't even parse it!" - Kibo, in ARK


  5. Re: Fuzzy Logic Operating Systems

    Ekrem SABAN wrote:
    > Hi everybody!
    >
    > I am currently a research associate at the Vienna University of Technology &
    > responsible for the creation of new projects. Currently, I have a new
    > subject that involves fuzzy control and fuzzy logic operating systems.
    >
    > What I want to discuss here is the reasons for using a fuzzy logic operating
    > system that makes some decisions using the theory of fuzzy logics.
    >

    Let me remark that the the things I _insist_ upon in a piece of computer
    hardware or operating system is predictability, repeatability, and
    deterministic behavior. If I do the same thing (e.g., run a program with the
    same data), I _absolutely insist_ I get the same answer. Similarly, if I run
    a suite of concurrent finite state machines that communicate with messages,
    I _absolutely insist_ that it work the same each time I run it. If I cannot
    have that, I cannot use it in a real-time situation.

    There are operating systems out there that are already non-deterministic and
    they are not worth a damn. To do it on purpose is a dirty trick, and this is
    not even April 1.

    --
    .~. Jean-David Beyer Registered Linux User 85642.
    /V\ PGP-Key: 9A2FC99A Registered Machine 241939.
    /( )\ Shrewsbury, New Jersey http://counter.li.org
    ^^-^^ 14:15:00 up 24 days, 5:43, 4 users, load average: 4.34, 4.24, 4.19

  6. Re: Fuzzy Logic Operating Systems

    On 2006-02-13, Jean-David Beyer wrote:
    > Ekrem SABAN wrote:
    >> Hi everybody!
    >>
    >> I am currently a research associate at the Vienna University of Technology &
    >> responsible for the creation of new projects. Currently, I have a new
    >> subject that involves fuzzy control and fuzzy logic operating systems.
    >>
    >> What I want to discuss here is the reasons for using a fuzzy logic operating
    >> system that makes some decisions using the theory of fuzzy logics.
    >>

    > Let me remark that the the things I _insist_ upon in a piece of computer
    > hardware or operating system is predictability, repeatability, and
    > deterministic behavior. If I do the same thing (e.g., run a program with the
    > same data), I _absolutely insist_ I get the same answer. Similarly, if I run
    > a suite of concurrent finite state machines that communicate with messages,
    > I _absolutely insist_ that it work the same each time I run it. If I cannot
    > have that, I cannot use it in a real-time situation.
    >
    > There are operating systems out there that are already non-deterministic and
    > they are not worth a damn. To do it on purpose is a dirty trick, and this is
    > not even April 1.


    Fuzzy logic is a misnomer, so don't extrapolate based on that
    name. It should really be called "the logic of fuzzy sets".

    --
    Chris F.A. Johnson, author |
    Shell Scripting Recipes: | My code in this post, if any,
    A Problem-Solution Approach | is released under the
    2005, Apress | GNU General Public Licence

  7. Re: Fuzzy Logic Operating Systems

    Chris F.A. Johnson wrote:
    > On 2006-02-13, Jean-David Beyer wrote:
    >> Ekrem SABAN wrote:
    >>> Hi everybody!
    >>>
    >>> I am currently a research associate at the Vienna University of Technology &
    >>> responsible for the creation of new projects. Currently, I have a new
    >>> subject that involves fuzzy control and fuzzy logic operating systems.
    >>>
    >>> What I want to discuss here is the reasons for using a fuzzy logic operating
    >>> system that makes some decisions using the theory of fuzzy logics.
    >>>

    >> Let me remark that the the things I _insist_ upon in a piece of computer
    >> hardware or operating system is predictability, repeatability, and
    >> deterministic behavior. If I do the same thing (e.g., run a program with the
    >> same data), I _absolutely insist_ I get the same answer. Similarly, if I run
    >> a suite of concurrent finite state machines that communicate with messages,
    >> I _absolutely insist_ that it work the same each time I run it. If I cannot
    >> have that, I cannot use it in a real-time situation.
    >>
    >> There are operating systems out there that are already non-deterministic and
    >> they are not worth a damn. To do it on purpose is a dirty trick, and this is
    >> not even April 1.

    >
    > Fuzzy logic is a misnomer, so don't extrapolate based on that
    > name. It should really be called "the logic of fuzzy sets".
    >

    I know that. I even met Lotfi Zadeh once, probably 30 years ago or so, when
    he was teaching at Princeton.

    --
    .~. Jean-David Beyer Registered Linux User 85642.
    /V\ PGP-Key: 9A2FC99A Registered Machine 241939.
    /( )\ Shrewsbury, New Jersey http://counter.li.org
    ^^-^^ 17:00:00 up 24 days, 8:28, 4 users, load average: 4.25, 4.19, 4.12

  8. Re: Fuzzy Logic Operating Systems

    On 2006-02-13, Jean-David Beyer wrote:
    > Ekrem SABAN wrote:
    >> Hi everybody!
    >>
    >> I am currently a research associate at the Vienna University of Technology &
    >> responsible for the creation of new projects. Currently, I have a new
    >> subject that involves fuzzy control and fuzzy logic operating systems.
    >>
    >> What I want to discuss here is the reasons for using a fuzzy logic operating
    >> system that makes some decisions using the theory of fuzzy logics.
    >>

    > Let me remark that the the things I _insist_ upon in a piece
    > of computer hardware or operating system is predictability,
    > repeatability, and deterministic behavior. If I do the same
    > thing (e.g., run a program with the same data), I _absolutely
    > insist_ I get the same answer.


    From what I recall from grad school, fuzzy logic meets that
    requirement.

    > Similarly, if I run a suite of concurrent finite state
    > machines that communicate with messages, I _absolutely insist_
    > that it work the same each time I run it.


    Again, I don't see why that wouldn't be the case with fuzzy
    logic.

    > If I cannot have that, I cannot use it in a real-time
    > situation.


    Fair enough.

    > There are operating systems out there that are already
    > non-deterministic and they are not worth a damn.


    Fuzzy logic isn't non-deterministic. At least not the way I
    remember it.

    > To do it on purpose is a dirty trick, and this is not even
    > April 1.


    --
    Grant Edwards grante Yow! Being a BALD HERO
    at is almost as FESTIVE as a
    visi.com TATTOOED KNOCKWURST.

  9. Re: Fuzzy Logic Operating Systems

    In comp.os.linux.misc Jean-David Beyer wrote:
    :>
    : Let me remark that the the things I _insist_ upon in a piece of computer
    : hardware or operating system is predictability, repeatability, and
    : deterministic behavior. If I do the same thing (e.g., run a program with the

    So you don't use any programs on binary computers that do floating
    point math? If so, ever heard of round-off error or wonder why
    5/5 is not necessarily equal to 2-1?

    Don't foget, ALL binary computers work in finite precision, so insist
    away!

    Stan

    --
    Stan Bischof ("stan" at the below domain)
    www.worldbadminton.com

  10. Re: Fuzzy Logic Operating Systems

    stan@worldbadminton.com wrote:
    > In comp.os.linux.misc Jean-David Beyer wrote:
    > :>
    > : Let me remark that the the things I _insist_ upon in a piece of computer
    > : hardware or operating system is predictability, repeatability, and
    > : deterministic behavior. If I do the same thing (e.g., run a program with the
    >
    > So you don't use any programs on binary computers that do floating
    > point math? If so, ever heard of round-off error or wonder why
    > 5/5 is not necessarily equal to 2-1?


    You are missing the point. What you get when doing floating point arithmetic
    on any given machine, with a given compilation system _is_ deterministically
    defined. So if floating point 5 divided by floating point 5 does not give
    exactly floating point zero (although in every case I know of, it does), and
    floating point 2 minus floating point 1 does not give exactly floating point
    1 (and it does in every case I know of), the floating point difference
    between the ratio and the difference, above, is _always the same_ or the
    machine is broken.
    >
    > Don't foget, ALL binary computers work in finite precision, so insist
    > away!
    >

    You do not get it.

    --
    .~. Jean-David Beyer Registered Linux User 85642.
    /V\ PGP-Key: 9A2FC99A Registered Machine 241939.
    /( )\ Shrewsbury, New Jersey http://counter.li.org
    ^^-^^ 18:20:00 up 24 days, 9:48, 4 users, load average: 4.15, 4.16, 4.16

  11. Re: Fuzzy Logic Operating Systems

    In comp.os.linux.misc Jean-David Beyer wrote:
    :>
    :> Don't foget, ALL binary computers work in finite precision, so insist
    :> away!
    :>
    : You do not get it.

    Sure- if you use exactly the same software running on exactly the same
    hardware starting from exactly the same initial conditions,
    and with all exactly the same inputs, in
    a deterministic system you end up in exactly the same ending point.

    However, that would apply to all binary computers running
    any OS/software, barring true random events like alpha-particle
    induced flipped bits and the like, no?

    can you name ANY OS/software that isn't determinitstic given this
    restrictivce definition?

    I was pointing out that if you are a little less restrictive then you end
    up with all sorts of affects - like FP bit errors- that creep in
    and appear to lend a randomness.

    regards

    Stan
    --
    Stan Bischof ("stan" at the below domain)
    www.worldbadminton.com

  12. Re: Fuzzy Logic Operating Systems

    >>>>> "Jean-David" == Jean-David Beyer writes:

    Jean-David> You are missing the point. What you get when doing
    Jean-David> floating point arithmetic on any given machine, with a
    Jean-David> given compilation system _is_ deterministically
    Jean-David> defined. So if floating point 5 divided by floating
    Jean-David> point 5 does not give exactly floating point zero
    Jean-David> (although in every case I know of, it does),

    In very case I know of, it doesn't. 5/5 != 0 That's simple arithmetic
    that one learns in primary school.


    Jean-David> and floating point 2 minus floating point 1 does not
    Jean-David> give exactly floating point 1 (and it does in every
    Jean-David> case I know of), the floating point difference between
    Jean-David> the ratio and the difference, above, is _always the
    Jean-David> same_ or the machine is broken.

    --
    Lee Sau Dan §õ¦u´° ~{@nJX6X~}

    E-mail: danlee@informatik.uni-freiburg.de
    Home page: http://www.informatik.uni-freiburg.de/~danlee

  13. Re: Fuzzy Logic Operating Systems

    >>>>> "Ekrem" == Ekrem SABAN writes:

    Ekrem> and the computer languages used to implement OSs are using
    Ekrem> crisp logic.

    Because that's sufficient for building systems that support fuzzy
    logic.

    Are you aware of CPUs that do complex numbers *natively*? And do you
    think that prevents computers from handling complex numbers?


    Ekrem> For instance, a C expression will either be matched (true)
    Ekrem> or not (false).

    That's not true. a C expression can be int, char, char*, ...


    Perhaps, you mean _regular expressions_? No, neither. A regular
    expression cannot be matched or unmatched, until you APPLY it on a
    particular TARGET string. The expression itself, standalone, is
    neither matched or unmatched.


    Ekrem> If this is the way OS have been designed, the question is
    Ekrem> if this the (best) right way to do so.



    --
    Lee Sau Dan §õ¦u´° ~{@nJX6X~}

    E-mail: danlee@informatik.uni-freiburg.de
    Home page: http://www.informatik.uni-freiburg.de/~danlee

  14. Re: Fuzzy Logic Operating Systems

    What do you mean with sufficient? You can write with an extensive assembler library everything. But is this the optimal way to do it? To know all the stuff you have to know to program in assembler? Or is it better to go to a higher level? The level you call sufficient, the crisp logic OS. I propose that another higher level would solve some problems before they even become problems. What I mean are those problems that arise when projecting the fuzzy world of human thinking to the simple but more exact world of crisp logic.

    Concerning C, you are right: there can be a lot of types to matched. But they are matched exactly. So, if I write

    if (strName = "anton" && fSalary > 1000.0) { ... }

    it will not match for "Anton" or "Antony", and it will for sure not take the "Anton" having a salary of 994.8, although it may be a POSSIBLE valid result, not PRECISE ENOUGH formulated for a crisp language, but humanly thinking perhaps too precise.

    "Lee Sau Dan" schrieb im Newsbeitrag news:871wy4k3ue.fsf@informatik.uni-freiburg.de...
    >>>>> "Ekrem" == Ekrem SABAN writes:


    Ekrem> and the computer languages used to implement OSs are using
    Ekrem> crisp logic.

    Because that's sufficient for building systems that support fuzzy
    logic.

    Are you aware of CPUs that do complex numbers *natively*? And do you
    think that prevents computers from handling complex numbers?


    Ekrem> For instance, a C expression will either be matched (true)
    Ekrem> or not (false).

    That's not true. a C expression can be int, char, char*, ...


    Perhaps, you mean _regular expressions_? No, neither. A regular
    expression cannot be matched or unmatched, until you APPLY it on a
    particular TARGET string. The expression itself, standalone, is
    neither matched or unmatched.


    Ekrem> If this is the way OS have been designed, the question is
    Ekrem> if this the (best) right way to do so.



    --
    Lee Sau Dan §õ¦u´° ~{@nJX6X~}

    E-mail: danlee@informatik.uni-freiburg.de
    Home page: http://www.informatik.uni-freiburg.de/~danlee

  15. Re: Fuzzy Logic Operating Systems

    Ekrem SABAN wrote:
    >
    > Part 1.1 Type: Plain Text (text/plain)
    > Encoding: quoted-printable


    Please do not use html or mime attachments in usenet. This is a
    pure text medium.

    --
    "If you want to post a followup via groups.google.com, don't use
    the broken "Reply" link at the bottom of the article. Click on
    "show options" at the top of the article, then click on the
    "Reply" at the bottom of the article headers." - Keith Thompson
    More details at:
    Also see



+ Reply to Thread