This action might not be possible to undo. Are you sure you want to continue?

LOGIC

History & applications

Fuzzy logic was first proposed by Lotfi A. Zadeh of the University of California at Berkeley in a 1965 paper. He elaborated on his ideas in a 1973 paper that introduced the concept of "linguistic variables", which in this article equates to a variable defined as a fuzzy set. Other research followed, with the first industrial application, a cement kiln built in Denmark, coming on line in 1975. Fuzzy systems were largely ignored in the U.S. because they were associated with artificial intelligence, a field that periodically oversells itself, especially in the mid-1980s, resulting in a lack of credibility within the commercial domain. The Japanese did not have this prejudice. Interest in fuzzy systems was sparked by Seiji Yasunobu and Soji Miyamoto of Hitachi, who in 1985 provided simulations that demonstrated the superiority of fuzzy control systems for the Sendai railway. Their ideas were adopted, and fuzzy systems were used to control accelerating, braking, and stopping when the line opened in 1987. Another event in 1987 helped promote interest in fuzzy systems. During an international meeting of fuzzy researchers in Tokyo that year, Takeshi Yamakawa demonstrated the use of fuzzy control, through a set of simple dedicated fuzzy logic chips, in an "inverted pendulum" experiment. This is a classic control problem, in which a vehicle tries to keep a pole mounted on its top by a hinge upright by moving back and forth. Observers were impressed with this demonstration, as well as later experiments by Yamakawa in which he mounted a wine glass containing water or even a live mouse to the top of the pendulum. The system maintained stability in both cases. Yamakawa eventually went on to organize his own fuzzy-systems research lab to help exploit his patents in the field. Following such demonstrations, Japanese engineers developed a wide range of fuzzy systems for both industrial and consumer applications. In 1988 Japan established the Laboratory for International Fuzzy Engineering (LIFE), a cooperative arrangement between 48 companies to pursue fuzzy research. Japanese consumer goods often incorporate fuzzy systems. Matsushita vacuum cleaners use microcontrollers running fuzzy algorithms to interrogate dust sensors and adjust suction power accordingly. Hitachi washing machines use fuzzy controllers to load-weight, fabric-mix, and dirt sensors and automatically set the wash cycle for the best use of power, water, and detergent. As a more specific example, Canon developed an autofocusing camera that uses a charge-coupled device (CCD) to measure the clarity of the image in six regions of its

field of view and use the information provided to determine if the image is in focus. It also tracks the rate of change of lens movement during focusing, and controls its speed to prevent overshoot. The camera's fuzzy control system uses 12 inputs: 6 to obtain the current clarity data provided by the CCD and 6 to measure the rate of change of lens movement. The output is the position of the lens. The fuzzy control system uses 13 rules and requires 1.1 kilobytes of memory. As another example of a practical system, an industrial air conditioner designed by Mitsubishi uses 25 heating rules and 25 cooling rules. A temperature sensor provides input, with control outputs fed to an inverter, a compressor valve, and a fan motor. Compared to the previous design, the fuzzy controller heats and cools five times faster, reduces power consumption by 24%, increases temperature stability by a factor of two, and uses fewer sensors. The enthusiasm of the Japanese for fuzzy logic is reflected in the wide range of other applications they have investigated or implemented: character and handwriting recognition; optical fuzzy systems; robots, including one for making Japanese flower arrangements; voice-controlled robot helicopters, this being no mean feat, as hovering is a "balancing act" rather similar to the inverted pendulum problem; control of flow of powders in film manufacture; elevator systems; and so on. Work on fuzzy systems is also proceeding in the US and Europe, though not with the same enthusiasm shown in Japan. The US Environmental Protection Agency has investigated fuzzy control for energy-efficient motors, and NASA has studied fuzzy control for automated space docking: simulations show that a fuzzy control system can greatly reduce fuel consumption. Firms such as Boeing, General Motors, Allen-Bradley, Chrysler, Eaton, and Whirlpool have worked on fuzzy logic for use in low-power refrigerators, improved automotive transmissions, and energy-efficient electric motors. In 1995 Maytag introduced an "intelligent" dishwasher based on a fuzzy controller and a "one-stop sensing module" that combines a thermistor, for temperature measurement; a conductivity sensor, to measure detergent level from the ions present in the wash; a turbidity sensor that measures scattered and transmitted light to measure the soiling of the wash; and a magnetostrictive sensor to read spin rate. The system determines the optimum wash cycle for any load to obtain the best results with the least amount of energy, detergent, and water. It even adjusts for dried-on foods by tracking the last time the door was opened, and estimates the number of dishes by the number of times the door was opened. Research and development is also continuing on fuzzy applications in software, as opposed to firmware, design, including fuzzy expert systems and integration of fuzzy logic with neural-network and so-called adaptive "genetic" software systems, with the ultimate goal of building "self-learning" fuzzy control systems.

and precise numeric values.INTRODUCTION This is the first in a series of six articles intended to share information and experience in the realm of fuzzy logic (FL) and its application. In consumer electronics. Fuzzy logic is a superset of conventional (Boolean) logic that has been extended to handle the concept of partial truth -. Nearly every application can potentially realize some of the benefits of Fuzzy Logic. Fuzzy Logic has been gaining increasing acceptance during the past few years. fuzzy logic resembles human decision making with its ability to work from approximate data and find precise solutions. In manufacturing. fuzzy improves time to market and helps reduce costs. In aerospace. such as performance. This article will introduce FL. Fuzzy logic incorporates an alternative way of thinking. Fuzzy provides a remarkably simple way to draw definite conclusions from vague. fuzzy is proven to be invaluable in increasing equipment efficiency and diagnosing malfunctions. (Note: Lotfi. There are over two thousand commercially available products using Fuzzy Logic. fuzzy enables very complex real time problems to be tackled using a simple approach. Unlike classical logic which requires a deep understanding of a system. ambiguous or imprecise information. Through the course of this article series. which allows modeling complex systems using a higher level of abstraction originating from our knowledge and experience. Several manufacturers in the automotive industry are using fuzzy technology to improve quality and reduce development time.) Zadeh says that rather than regarding fuzzy theory as a single theory. we should regard the process of ``fuzzification'' as a methodology to generalize ANY specific theory from a crisp (discrete) to a continuous .truth values between "completely true" and "completely false". ranging from washing machines to high speed trains. It was introduced by Dr. Lotfi Zadeh of UC/Berkeley in the 1960's as a means to model the uncertainty of natural language. simplicity. Fuzzy Logic allows expressing this knowledge with subjective concepts such as very hot. and a long time which are mapped into exact numeric ranges. bright red. not Lofti. What Is Fuzzy Logic ? Fuzzy logic is a powerful problem-solving methodology with a myriad of applications in embedded control and information processing. In a sense. a simple implementation will be explained in detail. and productivity. exact equations. lower cost. is the correct spelling of his name. Fuzzy Logic has been found to be very suitable for embedded control applications.

In MIN inferencing.(fuzzy) form (see "extension principle" in [2]). In PRODUCT inferencing. and medium is a membership function defined on z. z is an output variable (a name for a data value to be computed). while the conclusion (the rule's consequent) assigns a membership function to each of one or more output variables. instead of Boolean logic. to reason about data. to determine the degree of truth for each rule premise. the truth value for the premise of each rule is computed. Again. What is a fuzzy expert system? A fuzzy expert system is an expert system that uses a collection of fuzzy membership functions and rules. the combined output fuzzy subset is constructed by taking . The general inference process proceeds in three (or four) steps. "fuzzy differential equations". low is a membership function (fuzzy subset) defined on x. The antecedent (the rule's premise) describes to what degree the rule applies. Usually only MIN or PRODUCT are used as inference rules. Thus recently researchers have also introduced "fuzzy calculus". the output membership function is clipped off at a height corresponding to the rule premise's computed degree of truth (fuzzy logic AND). and so on . the output membership function is scaled by the rule premise's computed degree of truth. all of the fuzzy subsets assigned to each output variable are combined together to form a single fuzzy subset for each output variable. high is a membership function defined on y. Under FUZZIFICATION. and applied to the conclusion part of each rule. In MAX composition. Under INFERENCE. Most tools for working with fuzzy expert systems allow more than one conclusion per rule. 3. the membership functions defined on the input variables are applied to their actual values. 1. usually MAX or SUM are used. Under COMPOSITION. 2. The rules in a fuzzy expert system are usually of a form similar to the following: if x is low and y is high then z = medium where x and y are input variables (names for know data values). The set of rules in a fuzzy expert system is known as the rulebase or knowledge base. This results in one fuzzy subset to be assigned to each output variable for each rule.

the crisp value of the output variable is computed by finding the variable value of the center of gravity of the membership function for the fuzzy value. and z all take on values in the interval [0. Notes: 1.( t / 10 ) high(t) = t / 10 rule 1: if x is low and y is low then z is high rule 2: if x is low and y is high then z is low rule 3: if x is high and y is low then z is low rule 4: if x is high and y is high then z is high Notice that instead of assigning a single value to the output variable z. which is used when it is useful to convert the fuzzy output set to a crisp number. low(t)+high(t)=1. This is also not required. In the MAXIMUM method. the combined output fuzzy subset is constructed by taking the pointwise sum over all of the fuzzy subsets assigned to the output variable by the inference rule. but fairly common. This isn't required. This is not required. The value of t at which low(t) is maximum is the same as the value of t at which high(t) is minimum.the pointwise maximum over all of the fuzzy subsets assigned tovariable by the inference rule (fuzzy logic OR). Two of the more common techniques are the CENTROID and MAXIMUM methods. and vice-versa. and is also *not* common. and that the following membership functions and rules are defined: low(t) = 1 . 3. There are more defuzzification methods than you can shake a stick at (at least 30). In SUM composition.10]. 2. each rule assigns an entire fuzzy subset (low or high). In the CENTROID method. In the fuzzification subprocess. In this example. 4. the membership functions defined on the . y. but it is fairly common. The same membership functions are used for all variables. one of the variable values at which the fuzzy subset has its maximum truth value is chosen as the crisp value for the output variable.0 for all t. Extended Example: Assume that the variables x. Finally is the (optional) DEFUZZIFICATION.

0 0.0 0.0 0.32 0.0 1.2 3.068 * z Note: The terminology used here is slightly nonstandard.68 0.8 } For the same conditions. MIN and PRODUCT are two INFERENCE METHODS or INFERENCE RULES.0 1.39 0.0 0.0 0. In most texts. As shown in the table above. PRODUCT inferencing will assign z the fuzzy subset defined by the membership function: rule1(z) = 0.0 0.31 0.31 0.0 0..0 0.68 0.0 0.0 0.0 1.61 0.0 0.0 0.68 0.0 0. In PRODUCT inferencing. For example. the premise degree of truth works out to 0. .68. For example. the output membership function is clipped off at a height corresponding to the rule premise's computed degree of truth. If a rule's premise has a nonzero degree of truth (if the rule applies at all.67 0.61 0.0 0.0 0.33 0.3 0.0 6.0 0.0 0.33 0.0 1.32 0.0 0. MIN inferencing will assign z the fuzzy subset defined by the membership function: rule1(z) = { z / 10.2 0.0 0. let's look at rule 1 for x = 0. This results in one fuzzy subset to be assigned to each output variable for each rule.0 0.32 0.0 0..) then the rule is said to FIRE.0 0.68 0.0 1.0 0.0 0.0 0. In MIN inferencing.0 1.61 0.0 0.2. the output membership function is scaled by the rule premise's computed degree of truth.0 10.0 1.0 10.0 3.39 0. and applied to the conclusion part of each rule.0 0.0 0. if z >= 6.39 0.32 1.0 3.0 1.68 * high(z) = 0. if z <= 6.0 6.1 1.0 1.39 0.0 0.8 0.0 In the inference subprocess.32 0.32 0.0 0.68 0.1 0. The degree of truth for a rule's premise is sometimes referred to as its ALPHA. For this rule.0 0.61 1.0 0. the truth value for the premise of each rule is computed.69 0.0 0.0 0.0 1.68 0.32 0.32 0.0 1.67 0. x y low(x) high(x) low(y) high(y) alpha1 alpha2 alpha3 alpha4 -----------------------------------------------------------------------------0.0 0.68 0.0 and y = 3.68.0 0. to determine the degree of truth for each rule premise.2 3.0 0.0 0.0 0.2 1. This corresponds to the traditional interpretation of the fuzzy logic AND operation.input variables are applied to their actual values.0 10.1 0.0 3.32 10.0 1.31 3.

if z <= 6. all of the fuzzy subsets assigned to each output variable are combined together to form a single fuzzy subset for each output variable. the combined output fuzzy subset is constructed by taking the pointwise maximum over all of the fuzzy subsets assigned to the output variable by the inference rule.68. the combined output fuzzy subset is constructed by taking the pointwise sum over all of the fuzzy subsets assigned to the output variable by the inference rule.8 0." Thus you'll see such terms as "MAX-MIN inference" and "SUM-PRODUCT inference" in the literature. For example.8 } rule3(z) = 0. respectively. In SUM composition.2 z / 10. They are the combination of MAX composition and MIN inference.these mean the same things as the reverse order. or SUM composition and PRODUCT inference. You'll also see the reverse terms "MIN-MAX" and "PRODUCT-SUM" -.0 MAX composition would result in the fuzzy subset: fuzzy(z) = { 0.8 0. Note that this can result in truth values greater than one! For this reason. assume x = 0.68. if z >= 6.8 } . that doesn't have a problem with this odd case. In MAX composition. if z <= 6.0 rule4(z) = 0.z / 10.the term "inference method" is used to mean the combination of the things referred to separately here as "inference" and "composition.2 <= z <= 6. if z <= 3.32.8 } rule2(z) = { 0. It seems clearer to describe the two processes separately. SUM composition is only used when it will be followed by a defuzzification method.0 and y = 3. Otherwise SUM composition can be combined with normalization and is therefore a general purpose method again. if z >= 6.8 1 . MIN inferencing would assign the following four fuzzy subsets to z: rule1(z) = { z / 10. In the composition subprocess. such as the CENTROID method. if 3.32. if z >= 6.2. MAX composition and SUM composition are two COMPOSITION RULES.

A couple of years ago. but more often. Mizumoto did a short paper that compared about ten defuzzification methods.8) = 5. one of the variable values at which the fuzzy subset has its maximum truth value is chosen as the crisp value for the output variable.068 * z rule2(z) = 0. which is . go back to our previous examples. There are several variations of the MAXIMUM method that differ only in what they do when there is more than one variable value at which this maximum truth value occurs.32+0. you divide the moment of the function by the area of the function. To compute the moment of f(x). the AVERAGE-OF-MAXIMA method.036 * z Sometimes it is useful to just examine the fuzzy subsets that are the result of the composition process. Using MAX-MIN inferencing and AVERAGE-OF-MAXIMA defuzzification results in a crisp value of 8.32*z+0. In this case. we would compute the area as integral from 0 to 10 of (0. Two of the more common techniques are the CENTROID and MAXIMUM methods.018*100) = (3. you compute the integral of f(x) dx. 10]. In the MAXIMUM method.32 .0.2 + 1.036*z*z) dz.0 rule4(z) = 0. you compute the integral of x*f(x) dx.0 SUM composition would result in the fuzzy subset: fuzzy(z) = 0.32 + 0.32 * 10 + 0. This is what the defuzzification subprocess does. Using PRODUCT-SUM inferencing and CENTROID defuzzification results in a crisp value of 5.032 * z rule3(z) = 0.4 for z. we state that all variables (including z) take on values in the range [0. In the CENTROID method. as follows. the crisp value of the output variable is computed by finding the variable value of the center of gravity of the membership function for the fuzzy value. For example.6 for z. One of these. this FUZZY VALUE needs to be converted to a single number -. Earlier on in the FAQ. returns the average of the variable values at which the maximum truth value occurs.036*z) dz. and to compute the area of f(x). There are more defuzzification methods than you can shake a stick at. To compute the centroid of the function f(x).0 and the moment as the integral from 0 to 10 of (0.PRODUCT inferencing would assign the following four fuzzy subsets to z: rule1(z) = 0.a CRISP VALUE. which is (0.

electrical.012 * 10 * 10 * 10) = (16 + 12) = 28 Finally. What is fuzzy control? The purpose of control is to influence the behavior of a system by changing an input or inputs to that system according to a rule or set of rules that model how the system operates. reference----->0------->( SYSTEM ) -------+----------> output . The Sony PalmTop apparently uses a fuzzy logic decision tree algorithm to perform handwritten (well. Classic control theory uses a mathematical model to define a relationship that transforms the desired state (requested) and observed state (measured) of the system into an input or inputs that will alter the future state of that system. Where are fuzzy expert systems used? fuzzy expert systems are the most common use of fuzzy logic.(0. Most applications of fuzzy logic use it as the underlying logic system for fuzzy expert systems.6. chemical or any combination of these. including: o Linear and Nonlinear Control o Pattern Recognition o Financial Systems o Operation Research Where is fuzzy logic used? Fuzzy logic is used directly in very few applications. computer lightpen) Kanji character recognition. the centroid is 28/5 or 5. They are used in several wide-ranging fields.16 * 10 * 10 + 0. The system being controlled may be mechanical.

each of the form: . B and C are constants. the microcontroller then makes decisions for what action to take based on a set of "rules". This takes the output of the system and compares it with the desired state of the system. but the scheme can deal with them as simplified fuzzy functions that happen to be either one value or another. and such switch inputs of course will always have a truth value equal to either 1 or 0. INT(e)dt is the integral of the error over time and de/dt is the change in the error term. The process of inference binding them together to produce the desired outputs.e + B. e is the error term.INT(e)dt + C. known as "fuzzy sets". a fuzzy model has replaced the mathematical one. As the complexity of the system increases it becomes more difficult to formulate that mathematical model. inputs along with its analog inputs. The inputs and outputs of the system have remained unchanged. A control system may also have various types of switch. A. or "ON-OFF". output = A. The Sendai subway is the prototypical example application of fuzzy control. The process of converting a crisp input value to a fuzzy value is called "fuzzification". It adjusts the input value based on the difference between the two values according to the following equation.de/dt Where. the role of the mathematical model and replaces it with another that is build from a number of smaller rules that in general only describe a small section of the whole system. Fuzzy sets The input variables in a fuzzy control system are in general mapped into by sets of membership functions similar to this. Given "mappings" of input variables into membership functions and truth values.^ | | | +--------( MODEL )<--------+feedback The most common example of a control model is the PID (proportional-integralderivative) controller. in the picture above. The major drawback of this system is that it usually assumes that the system being modelled in linear or at least behaves in some fashion that is a monotonic function. Fuzzy control replaces. That is.

In many cases. In any case there is sometime of . and so on. the mathematical model of the control process may not exist. fuzzy logic is well suited to low-cost implementations based on cheap sensors. Such systems can be easily upgraded by adding new rules to improve performance or add new features. As an example. is also defined by a fuzzy set that can have values like "static". if we consider systems of rules in which the class antecedent define a partition such a paradoxical phenomenon does not arise. If PID and other traditional control systems are so well-developed. Obviously. why bother with fuzzy control? It has some advantages. "slightly decreased". or may be too "expensive" in terms of computer processing power and memory. Traditional control systems are based on mathematical models in which the control system is described using one or more differential equations that define the system response to its inputs. and 4-bit or 8-bit one-chip microcontroller chips. Then the formula Cold(r)→High(t) is true for any t and therefore any t gives a correct control given r. and a system based on empirical rules may be more effective. the actual brake pressure. to determine the result of the rule. "slightly increased". and are highly effective. In this example. In many cases. Logical interpretation of fuzzy control In spite of the appearance there are several difficulties to give a rigorous logical interpretation of the IF-THEN rules. Furthermore. using the membership functions and truth values obtained from the inputs. These results are combined to give a specific ("crisp") answer.IF brake temperature IS warm AND speed IS not very fast THEN brake pressure IS slightly decreased. Such systems are often implemented as "PID controllers" (proportional-integral-derivative controllers). This combination of fuzzy operations and rule-based "inference" describes a "fuzzy expert system". This result in turn will be mapped into a membership function and truth value controlling the output variable. "brake pressure". They are the products of decades of development and theoretical analysis. This rule by itself is very puzzling since it looks like it could be used without bothering with fuzzy logic. low-resolution analog-to-digital converters. The output variable. fuzzy control can be used to improve existing traditional controller systems by adding an extra layer of intelligence to the current control method. but remember the decision is based on a set of rules: • • • All the rules that apply are invoked. the two input variables are "brake temperature" and "speed" that have values defined as fuzzy sets. interpret a rule as IF (temperature is "cold") THEN (heater is "high") by the first order formula Cold(x)→High(y) and assume that r is an input such that Cold(r) is false. a procedure known as "defuzzification".

we interpret f(r.y) in the associated least fuzzy Herbrand model.e. It was developed by the Integrated Reasoning Group of the Institute for Information Technology of the National Research Council of Canada and has been widely distributed for a number of years. In practice. A rigorous logical justification of fuzzy control is given in Hájek's book (see Chapter 7) where fuzzy control is represented as a theory of Hájek's basic logic. i.t) as the truth degree of the claim "t is a good answer given r". Neurofuzzy system (the more popular term is used henceforth) incorporates the human-like reasoning style of fuzzy systems through the use of fuzzy sets and a linguistic model consisting of a set of IF-THEN fuzzy rules. one of the two properties prevails. The strength of neuro-fuzzy systems involves two contradictory requirements in fuzzy modeling: interpretability versus accuracy.y) is the fuzzy set of possible outputs. Also in Gerla 2005 a logical approach to fuzzy control is proposed based on the following idea. The neuro-fuzzy in fuzzy modeling research field is divided into two areas: . given the input r. Neuro-fuzzy was proposed by J. Then given a possible output 't'. s(y) = f(r. More formally. The system uses two basic inexact concepts.. Jang. neuro-fuzzy refers to combinations of artificial neural networks and fuzzy logic. FuzzyCLIPS FuzzyCLIPS is a fuzzy logic extension of the CLIPS (C Language Integrated Production System) expert system shell from NASA. The learning process becomes a question belonging to inductive logic theory. fuzzy (or inexact). Neuro-fuzzy hybridization is widely termed as Fuzzy Neural Network (FNN) or Neuro-Fuzzy System (NFS) in the literature. Denote by f the fuzzy function associated with the fuzzy control system.unsatisfactory in considering two variables x and y in a rule without some kind of functional dependence. and combined reasoning. It has provided a useful environment for developing fuzzy applications but it does require significant effort to update and maintain as new versions of CLIPS are released. It enhances CLIPS by providing a fuzzy reasoning capability that is fully integrated with CLIPS facts and inference engine allowing one to represent and manipulate fuzzy facts and rules. Neuro-fuzzy In the field of artificial intelligence. In such a way fuzzy control becomes a chapter of fuzzy logic programming. R. any system of IF-THEN rules can be translate into a fuzzy program in such a way that the fuzzy function f is the interpretation of a vague predicate Good(x. The main strength of neuro-fuzzy systems is that they are universal approximators with the ability to solicit interpretable IF-THEN rules. fuzziness and uncertainty. FuzzyCLIPS can deal with exact. Neuro-fuzzy hybridization results in a hybrid intelligent system that synergizes these two techniques by combining the human-like reasoning style of fuzzy systems with the learning and connectionist structure of neural networks. allowing fuzzy and normal terms to be freely mixed in the rules and facts of an expert system. S.

fuzzy inference and defuzzification through multilayers feed-forward connectionist networks.[ Pseudo outer-product-based fuzzy neural networks Pseudo outer-product-based fuzzy neural networks ("POPFNN") are a family of neurofuzzy systems that are based on the linguistic fuzzy model. Although generally assumed to be the realization of a fuzzy system through connectionist networks. Fuzzy membership generation 2. mainly the Mamdani model. Fuzzy logic based tuning of neural network training parameters. It must be pointed out that interpretability of the Mamdani-type neuro-fuzzy systems can be lost. condition and consequence layers. which is based on Truth Value Restriction The "POPFNN" architecture is a five-layer neural network where the layers from 1 to 5 are called: input linguistic layer. The fuzzification of the inputs and the defuzzification of the outputs are respectively performed by the input linguistic and output linguistic layers while the fuzzy inference is collectively performed by the rule. To improve the interpretability of neuro-fuzzy systems. Supervised fine-tuning . Representing fuzzification. which is based on commonly accepted fuzzy Compositional Rule of Inference[4] POPFNN-TVR. which is based on the Approximate Analogical Reasoning Scheme[3] POPFNN-CRI(S).linguistic fuzzy modeling that is focused on interpretability. rule layer. Fuzzy logic criteria for increasing a network size. mainly the Takagi-SugenoKang (TSK) model. output linguistic layer. this term is also used to describe some other configurations including: • • • • • Deriving fuzzy rules from trained RBF networks. The learning process of POPFNN consists of three phases: 1. certain measures must be taken. consequent layer. Realising fuzzy membership function through clustering algorithms in unsupervised learning in SOMs and neural networks. and precise fuzzy modeling that is focused on accuracy. wherein important aspects of interpretability of neuro-fuzzy systems are also discussed.[2] Three members of POPFNN exist in the literature: • • • POPFNN-AARS(S). Fuzzy rule identification 3. condition layer.

as of early 2009. which is analogous to probability reducing to determinism when unpredictability vanishes. where the third dimension is the value of the membership function at each point on its two-dimensional domain that is called its footprint of uncertainty (FOU). whereas a general type-2 . In order to symbolically distinguish between a type-1 fuzzy set and a type-2 fuzzy set. the first of which he called a type-2 fuzzy set. whereas Ã denotes the comparable type-2 fuzzy set. since that word has the connotation of lots of uncertainty. Ã. 1). 2. The present article focuses only on type-2 fuzzy sets because they are the next step in the logical progression from type-1 to typen fuzzy sets. A denotes a type-1 fuzzy set. The membership function of a general type-2 fuzzy set. Prof. something that seems to contradict the word fuzzy. Zadeh didn’t stop with type-2 fuzzy sets. So. So. then a type-2 fuzzy set reduces to a type-1 fuzzy set. a tilde symbol is put over the symbol for the fuzzy set. Zadeh [27]. and is a way to address the above criticism of type-1 fuzzy sets head-on. the resulting type-2 fuzzy set is called a general type-2 fuzzy set (to distinguish it from the special interval type-2 fuzzy set). For an interval type-2 fuzzy set that third-dimension value is the same (e. Type-2 fuzzy sets and systems Type-2 fuzzy sets and systems generalize (type-1) fuzzy sets and systems so that more uncertainty can be handled. A type-2 fuzzy set lets us incorporate uncertainty about the membership function into fuzzy set theory. they are the same fuzzy set). 1) everywhere. which means that no new information is contained in the third dimension of an interval type-2 fuzzy set. And. Fuzzy Kohonen Partitioning (FKP) or Discrete Incremental Clustering (DIC). the third dimension is ignored. Prof. where n = 1. if there is no uncertainty. When the latter is done. and only the FOU is used to describe it. criticism was made about the fact that the membership function of a type-1 fuzzy set has no uncertainty associated with it. It is for this reason that an interval type-2 fuzzy set is sometimes called a first-order uncertainty fuzzy set model. so.Various fuzzy membership generation algorithms can be used: Learning Vector Quantization (LVQ). Generally. what does one do when there is uncertainty about the value of the membership function? The answer to this question was provided in 1975 by the inventor of fuzzy sets. is three-dimensional (Fig. because in that 1976 paper [27] he also generalized all of this to type-n fuzzy sets. Although some researchers are beginning to explore higher than type-2 fuzzy sets.. this work is in its infancy. In the following discussions it is names may be used. … .g. for such a set. Lotfi A. the POP algorithm and its variant LazyPOP are used to identify the fuzzy rules. From the very beginning of fuzzy sets. when he proposed more sophisticated kinds of fuzzy sets.

The FOU represents the blurring of a type-1 membership function. variance and skewness [22] and uncertainty bounds [26] Similarity ([1]. [25]) Subsethood [21] Fuzzy set ranking [25] Fuzzy rule ranking and selection [31] Type-reduction methods ([7]. so even though Zadeh proposed type-2 fuzzy sets in 1976. [12]) Other uncertainty measures [fuzziness. Both kinds of fuzzy sets are being actively researched by an evergrowing number of researchers around the world. whereas the literature about general type-2 fuzzy sets is much smaller. [12]) Fuzzy weighted average [9] Linguistic weighted average [23] Synthesizing an FOU from data that are collected from a group of subject [10] . Formulas for the following have already been worked out for interval type-2 fuzzy sets: • • • • • • • • • • • • Fuzzy set operations: union. cardinality. the literature about interval type-2 fuzzy sets is large. People were still trying to figure out what to do with type-1 fuzzy sets. Interval Type-2 Fuzzy Sets Interval type-2 fuzzy sets have received the most attention because the mathematics that is needed for such sets—primarily interval arithmetic—is much simpler than the mathematics that is needed for general type-2 fuzzy sets. Since then. This means that engineers and scientists who already know type-1 fuzzy sets will not have to invest a lot of time learning about general type-2 fuzzy set mathematics in order to understand and use interval type-2 fuzzy sets. both of which are type-1 fuzzy sets! Consequently. Jerry Mendel and his student’s works on type-2 fuzzy sets and systems . [12]) Firing intervals for an interval type-2 fuzzy logic system ([3]. it is possible to use type-1 fuzzy set mathematics to characterize and work with interval type-2 fuzzy sets. and is completely described by its two bounding functions . [12]) Centroid (a very widely used operation by practitioners of such sets. although a small number of articles were published about them. a lower membership function (LMF) and an upper membership function (UMF). the time was not right for researchers to drop what they were doing with type-1 fuzzy sets to focus on type-2 fuzzy sets. and also an important uncertainty measure for them) ([7]. Work on type-2 fuzzy sets languished during the 1980’s and early-to-mid 1990’s. intersection and complement ([6]. [24]. [8]. So. more and more researchers around the world are writing articles about type-2 fuzzy sets and systems.fuzzy set (with its useful third-dimension) is sometimes referred to as a second-order uncertainty fuzzy set model. This changed in the latter part of the 1990’s as a result of Prof.

Frequently.1]. 1}. and the second element from the interval [0. The set S is referred to as the UNIVERSE OF DISCOURSE for the fuzzy subset F. In classical set theory. the mapping is described as a function. In practice. . The truth or falsity of the statement x is in U is determined by finding the ordered pair whose first element is x. 1}. each with the first element from S. with exactly one ordered pair present for each element of S. This defines a mapping between elements of the set S and values in the interval [0. 1} This mapping may be represented as a set of ordered pairs. and the second element is an element of the set {0.Fuzzy Subsets: Just as there is a strong relationship between Boolean logic and the concept of a subset. a fuzzy subset F of a set S can be defined as a set of ordered pairs. and the statement is false if it is 0. the MEMBERSHIP FUNCTION of F. The degree to which the statement x is in F is true is determined by finding the ordered pair whose first element is x. a subset U of a set S can be defined as a mapping from the elements of S to the elements of the set {0. The first element of the ordered pair is an element of the set S. the terms "membership function" and fuzzy subset get used interchangeably. The DEGREE OF TRUTH of the statement is the second element of the ordered pair. U: S --> {0. Similarly. there is a similar strong relationship between fuzzy logic and fuzzy subset theory. and values in between are used to represent intermediate DEGREES OF MEMBERSHIP. The value zero is used to represent complete non-membership. and the value one is used to represent membership. with exactly one ordered pair present for each element of S. the value one is used to represent complete membership. The value zero is used to represent non-membership.1]. The statement is true if the second element of the ordered pair is 1.

)/2ft. "Drew is TALL" = 0.. } A graph of this looks like: 1.38 Erik 5' 10" 0.. Let's talk about people and "tallness". At minimum. if height(x) < 5 ft. -> Given this definition.42 Mark 6' 1" 0. tall(x) = { 0.38.0 7. which represents our cognitive category of "tallness". if height(x) > 7 ft. they tend to be triangles pointing up.54 Kareem 7' 2" 1.00 [depends on who you ask] Expressions like "A is X" can be interpreted as degrees of truth.0 height. the discussion characterizes membership functions as if they always are based on a . In this case the set S (the universe of discourse) is the set of people. Also.00 [I think] Yoke 5' 5" 0. which will answer the question "to what degree is person x tall?" Zadeh describes TALL as a LINGUISTIC VARIABLE. here are some example values: Person Height degree of tallness -------------------------------------Billy 3' 2" 0.0 +-------------+-----+------------------| | 5. so here's an example.g. <= height (x) <= 7 ft. To each person in the universe of discourse. and they can be much more complex than that. if 5 ft. we have to assign a degree of membership in the fuzzy subset TALL. (height(x)-5ft. Note: Membership functions used in most applications almost never have as simple a shape as tall(x). The easiest way to do this is with a membership function based on the person's height. ft. Let's define a fuzzy subset TALL.21 Drew 5' 9" 0. 1.0 + +------------------| / | / 0...5 + / | / | / 0. e.That's a lot of mathematical baggage.

how do we interpret a statement like X is LOW and Y is HIGH or (not Z is MEDIUM) The standard definitions in fuzzy logic are: truth (not x) = 1. It's also possible to have even more criteria. This effectively establishes fuzzy subsets and logic as a true generalization of classical set theory and logic. 1}. and there is no conflict between fuzzy and crisp methods. or to have the membership function depend on elements from two completely different universes of discourse..assume the same definition of TALL as above. if age(x) > 60 yr. It's referred to as a two-dimensional membership function. if 18 yr. for example. assume that we have a fuzzy subset OLD defined by the membership function: old (x) = { 0. and occasionally used in practice.)/42 yr.single criterion. or a "fuzzy relation". although it is quite common. let . Some examples -. 1. In fact. Logic Operations: Now that we know what a statement like "X is LOW" means in fuzzy logic. truth(y)) truth (x or y) = maximum (truth(x). Note that if you plug just the values zero and one into these definitions. This is known as the EXTENSION PRINCIPLE. by this reasoning all crisp (traditional) subsets ARE fuzzy subsets of this very special type. } And for compactness. One could. and in addition. you get the same truth tables as you would expect from conventional Boolean logic.0 .truth (x) truth (x and y) = minimum (truth(x). if age(x) < 18 yr. which states that the classical results of Boolean logic are recovered from fuzzy logic operations when all fuzzy membership grades are restricted to the traditional set {0. <= age(x) <= 60 yr. (age(x)-18 yr. This is perfectly legitimate. but the definition for the NOT operation seems to be safe. want to have the membership function for TALL depend on both a person's height and their age (he's tall for his age). truth(y)) Some researchers in fuzzy logic have explored the use of other interpretations of the AND and OR operations. but this isn't always the case.

21 0. The reason for which is very simple.21 0.00 0. expert systems) in system design.00 1.a = X is TALL and X is OLD b = X is TALL or X is OLD c = not (X is TALL) Then we can compute the following values.79 5' 9" 27 0.00 0.00 0. and purely logic-based approaches (e. fuzzy design can accommodate the ambiguities of real-world human language and logic.54 0.9652 3' 4" 1. linear control design).29 0.62 5' 10" 32 0.00 1.g.21 0.00 0.64 0. While other approaches require accurate equations to model real-world behaviors.7780 6' 1" 1.00 0.38 0. here's a dandy little conversion table: Feet+Inches = Meters -------------------3' 2" 0.33 0. What does it offer? .42 0.g. height age X is TALL X is OLD a b c -----------------------------------------------------------------------3' 2" 65 0.31 0.00 5' 5" 30 0.00 For those of you who only grok the metric system.1844 Fuzzy logic has rapidly become one of the most successful of today's technologies for developing sophisticated control systems.6510 5' 9" 1.54 0.33 0.7526 5' 10" 1.58 6' 1" 31 0.29 0.00 1.0160 5' 5" 1.21 0. It fills an important gap in engineering design methods left vacant by purely mathematical approaches (e.00 1.8542 7' 2" 2.64 1.00 0.46 7' 2" 45 1.31 0.00 3' 4" 4 0.38 0. It provides both an intuitive method for describing systems in human terms and automates the conversion of those system specifications into effective models. Fuzzy logic addresses such applications perfectly as it resembles human decision making with an ability to generate precise solutions from certain or approximate information.42 0.

Fuzzy logic corresponds to "degrees of truth". and second. The meaning of each of them can be represented by a certain fuzzy set. The fuzzy car. In 1987. manuvers itself by following simple verbal instructions from its driver. these are simply different concepts which superficially seem similar because of using the same unit interval of real numbers [0. Then one might define the glass as being 0. Another designer might equally well design a set membership function where the glass would be considered full for all values down to 50 ml. Degrees of truth Fuzzy logic and probabilistic logic are mathematically similar – both have truth values ranging between 0 and 1 – but conceptually distinct. that after a few minutes. potatoes. Place chili. fuzzy-logic controllers make subway journeys more comfortable with smooth braking and acceleration. let a 100 ml glass contain 30 ml of water. e. automatically choosing the best cycle. affecting things we use everyday. fuzzy logic was used in more useful applications. and it cooks for the right time at the proper temperature. But. due to different interpretations -see interpretations of probability theory. practically the most exciting thing about it. Take for example.3 full. Still. the first fuzzy logic-controlled subway was opened in Sendai in northern Japan. all the driver has to do is push the start button! Fuzzy logic was also put to work in elevators to reduce waiting time. probability has nothing in common with fuzziness. while probabilistic logic corresponds to "probability. as the technology was further embraced. or both. Note that the conditioning can be achieved by having a specific observer that randomly selects the level for the glass. the fuzzy washing machine . the applications of Fuzzy Logic technology have virtually exploded.g. has no sense without accepting occurrence of some event. For example. It is essential to realize that fuzzy logic uses truth degrees as a mathematical model of the vagueness phenomenon while probability is a mathematical model of randomness. Since then. It can even stop itself when there is an obstacle immedeately ahead using sensors.1]. The fuzzy microwave. Best of all. likelihood". conditional distributions describing the probability that someone would call the glass full given a specific fullness level. Consequently. one can see where the confusion might arise. Both degrees of truth and probabilities range between 0 and 1 and hence may seem similar at first.7 empty and 0. Then we may consider two concepts: Empty and Full. as these differ. A load of clothes in it and press start. since theorems such as De Morgan's have dual applicability and properties of random variables are analogous to properties of binary logic states. or etc in a fuzzy microwave and push single button. the glass will be half empty. Note that the concept of emptiness would be subjective and thus would depend on the observer or designer.The first applications of fuzzy theory were primaly industrial. fuzzy logic and probabilistic logic yield different models of the same real-world situations. This model. a distribution over deterministic observers. . However. however. A probabilistic setting would first define a scalar variable for the fullness of the glass. is the simplicity involved in operating it. such as process control for cement kilns. and the machine begins to churn. Here.

the great utility of linguistic variables is that they can be modified via linguistic hedges applied to primary terms. A. Example Fuzzy set theory defines fuzzy operators on fuzzy sets. OR. and complement. Linguistic variables While variables in mathematics usually take numerical values.[4] A linguistic variable such as age may have a value such as young or its antonym old. the non-numeric linguistic variables are often used to facilitate the expression of rules and facts. maximum. Rules are usually expressed in the form: IF variable IS property THEN action For example. usually defined as the minimum. Each function maps the same temperature value to a truth value in the 0 to 1 range. does not work properly. fuzzy logic usually uses IF-THEN rules. because the temperature might be "cold" and "normal" at the same time to different degrees. For this reason. The linguistic hedges can be associated with certain functions. However. So for the fuzzy variables x and y: . This model. in fuzzy logic applications. L. a simple temperature regulator that uses a fan might look like this: IF temperature IS very cold THEN stop fan IF temperature IS cold THEN turn down fan IF temperature IS normal THEN maintain level IF temperature IS hot THEN speed up fan There is no "ELSE" – all of the rules are evaluated. For example. they are called the Zadeh operators. a temperature measurement for anti-lock brakes might have several separate membership functions defining particular temperature ranges needed to control the brakes properly. These truth values can then be used to determine how the brakes should be controlled. however. or constructs that are equivalent.Applying truth values A basic application might characterize subranges of a continuous variable. Zadeh proposed to take the square of the membership function. For instance. such as fuzzy associative matrices. and NOT operators of boolean logic exist in fuzzy logic. when they are defined this way. The problem in applying this is that the appropriate fuzzy operator may not be known. The AND.

These are generally adverbs such as "very".truth(x)) x AND y = minimum(truth(x). and implication is defined as the residuum of the t-norm. Fuzzy logic with evaluated syntax (sometimes also called Pavelka's logic). called hedges that can be applied. more linguistic in nature. A generalization of classical Gödel completeness theorem is provable in EVŁ. Propositional fuzzy logics The most important propositional fuzzy logics are: • • • • • • Monoidal t-norm-based propositional fuzzy logic MTL is an axiomatization of logic where conjunction is defined by a left continuous t-norm. This means that each formula has an evaluation. While the above kinds of fuzzy logic have traditional syntax and many-valued semantics. Łukasiewicz fuzzy logic is the extension of basic fuzzy logic BL where standard conjunction is the Łukasiewicz t-norm. and implication is also defined as the residuum of the t-norm.NOT x = (1 . It has the axioms of basic fuzzy logic plus an axiom of double negation. Product fuzzy logic is the extension of basic fuzzy logic BL where conjunction is product t-norm. and its models correspond to MV-algebras. truth(y)) x OR y = maximum(truth(x). in EVŁ is evaluated also syntax. and its models are called G-algebras. supremum) of the truth degrees of the instances of the quantified subformula. denoted by EVŁ. Predicate fuzzy logics These extend the above-mentioned fuzzy logics by adding universal and existential quantifiers in a manner similar to the way that predicate logic is created from propositional logic. and its models are called product algebras. It has the axioms of BL plus another axiom for cancellativity of conjunction. existential) quantifier in t-norm fuzzy logics is the infimum (resp. Its models correspond to MTL-algebras that are prelinear commutative bounded integral residuated lattices. It has the axioms of BL plus an axiom of idempotence of conjunction. which modify the meaning of a set using a mathematical formula. Basic propositional fuzzy logic BL is an extension of MTL logic where conjunction is defined by a continuous t-norm. truth(y)) There are also other operators. Axiomatization of EVŁ stems from Łukasziewicz fuzzy logic. is a further generalization of mathematical fuzzy logic. The semantics of the universal (resp. or "somewhat". Its models correspond to BL-algebras. Gödel fuzzy logic is the extension of basic fuzzy logic BL where conjunction is Gödel t-norm. .

A. Vila et al. the question of a suitable extension of such concepts to fuzzy set theory arises. the following theorem holds true (provided that the deduction apparatus of the fuzzy logic satisfies some obvious effectiveness property).M. fuzzy thresholds. Medina. Later.Decidability issues for fuzzy logic The notions of a "decidable subset" and "recursively enumerable subset" are basic ones for classical mathematics and classical logic. Application areas . the function h(x. Successively. Santos by the notions of fuzzy Turing machine. To this aim. Ü denotes the set of rational numbers in [0. Indeed. An extension of such a theory to the general case of the L-subsets is proposed in Gerla 2006. The first fuzzy relational database.1]. some other models arose like the Buckles-Petry model.n) is increasing with respect to n and s(x) = lim h(x. L. Another open question is to start from this notion to find an extension of Gödel’s theorems to fuzzy logic. some fuzzy querying languages have been defined. Fuzzy databases Once fuzzy relations are defined. Gerla showed that such a definition is not adequate and therefore proposed the following one. Biacino and G. A fuzzy subset s : S [0. further investigations on the notions of fuzzy grammar and fuzzy Turing machine should be necessary (see for example Wiedermann's paper). the fuzzy set of logically true formulas is recursively enumerable in spite of the fact that the crisp set of valid formulas is not recursively enumerable. like fuzzy conditions. linguistic labels and so on. In particular. fuzzy constants.1] of a set S is recursively enumerable if a recursive map h : S×N Ü exists such that. It is an open question to give supports for a Church thesis for fuzzy logic claiming that the proposed notion of recursive enumerability for fuzzy subsets is the adequate one. appeared in Maria Zemankova's dissertation. Theorem. Bosc et al. Galindo et al. Any axiomatizable fuzzy theory is recursively enumerable. A first proposal in such a direction was made by E. fuzzy constraints. it is possible to develop fuzzy relational databases.S. the Umano-Fukami model or the GEFRED model by J. the Prade-Testemale Model. FRDB. These languages define some structures in order to include fuzzy aspects in the SQL statements. highlighting the SQLf by P. for every x in S. fuzzy comparators. We say that s is decidable if both s and its complement –s are recursively enumerable. and the FSQL by J. any axiomatizable and complete theory is decidable. Markov normal fuzzy algorithm and fuzzy program (see Santos 1970). M. in general. Moreover. Then. In the context of fuzzy databases.n). The proposed definitions are well related with fuzzy logic.

the fuzzy-logicderived possibility measure is inherently different from the probability measure.e. how probable do I think that a variable is in a set). This approach to set theory was not applied to control systems until the 70's due to insufficient small-computer capability prior to that time. While both fuzzy logic and probability theory can be used to represent subjective belief... On the other hand. as probability only handles one kind of uncertainty. how much a variable is in a set). washing machine) Comparison to probability Fuzzy logic and probability are different ways of expressing uncertainty. However. While this distinction is mostly philosophical. Lotfi Zadeh argues that fuzzy logic is different in character from probability. He also claims to have proven a derivation of Bayes' theorem from the concept of fuzzy subsethood. Bart Kosko argues that probability is a subtheory of fuzzy logic. Freescale 68HC12) Hydrometeor classification algorithms for polarimetric weather radar Language filters on message boards and chat rooms for filtering out offensive text The Massive engine used in the Lord of the Rings films. probability theory uses the concept of subjective probability (i. many statisticians are persuaded by the work of Bruno de Finetti that only one kind of mathematical uncertainty is needed and thus fuzzy logic is unnecessary.Fuzzy logic is used in the operation or programming of: • • • • • • • • • • • • • • • • Air conditioners Automobile and such vehicle subsystems as automatic transmissions. hence they are not directly equivalent. a professor at the University of California at Berkley. such as edge detection Dishwashers Elevators Some microcontrollers and microprocessors (e. . which allowed largescale armies to enact random yet orderly movements Mineral Deposit estimation Pattern recognition in Remote Sensing Rice cookers Video game artificial intelligence Home appliances (e. fuzzy set theory uses the concept of fuzzy set membership (i. ABS and cruise control Tokyo monorail Cameras Digital image processing. but as a way of processing data by allowing partial set membership rather than crisp set membership or non-membership.e. and presented not as a control methodology. WHERE DID FUZZY LOGIC COME FROM? The concept of Fuzzy Logic (FL) was conceived by Lotfi Zadeh. He fuzzified probability to fuzzy probability and also generalized it to what is called possibility theory.g.g. and is not a replacement for it.

For example.S. manufacturers have not been so quick to embrace this technology while the Europeans and Japanese have been aggressively building real products around it. HOW DOES FL WORK? FL requires some numerical parameters in order to operate such as what is considered significant error and significant rate-of-change-of-error. FL is so forgiving that the system will probably work the first time without any tweaking. over which the members of the universe are completely in the set. If feedback controllers could be programmed to accept noisy. numerical information input. A typical . terms like "IF (process is too cool) AND (process is getting colder) THEN (add heat to the process)" or "IF (process is too hot) AND (process is heating rapidly) THEN (cool the process quickly)" are used. What are fuzzy numbers and fuzzy arithmetic? Fuzzy numbers are fuzzy subsets of the real line. rule-based IF X AND Y THEN Z approach to a solving control problem rather than attempting to model a system mathematically. or "210C <TEMP <220C". hereafter called "error-dot". The "error-dot" might then have units of degs/min with a small error-dot being 5F/min and a large one being 15F/min.Professor Zadeh reasoned that people do not require precise. imprecise input. rather than dealing with temperature control in terms such as "SP =500F". Generally. Fuzzy numbers are used very widely in fuzzy control applications. relying on an operator's experience rather than their technical understanding of the system. FL is capable of mimicking this type of behavior but at very high rate. they would be much more effective and perhaps easier to implement. The FL model is empirically-based. Unfortunately. The membership function is increasing towards the peak and decreasing away from it. HOW IS FL DIFFERENT FROM CONVENTIONAL CONTROL METHODS? FL incorporates a simple. "T <1000F". These terms are imprecise and yet very descriptive of what must actually happen. a simple temperature control system could use a single temperature feedback sensor whose data is subtracted from the command signal to compute "error" and then time-differentiated to yield the error slope or rate-of-change-of-error. but exact values of these numbers are usually not critical unless very responsive performance is required in which case empirical tuning would determine them. Error might have units of degs F and a small error considered to be 2F while a large error is 5F. For example. They have a peak or plateau with membership grade 1. Consider what you do in the shower if the temperature is too cold: you will make the water comfortable very quickly with little trouble. U. and yet they are capable of highly adaptive control. These values don't have to be symmetrical and can be "tweaked" once the system is operating in order to optimize performance.

At they very least.0 +-------------+-----+-----+-------------| | | 5. for example to choosing just the central value and the slope on either side.0 + + | /\ | / \ 0. as are exponential curves similar to Gaussian probability densities. There are a variety of possible conversion methods. Ad-hoc forms While there is a vast (hugely infinite) array of possible membership function forms. Slope and trapezoidal functions are also used. users can be tested using psychological methods. However. for example simple forms of fuzzy numbers (see [7]). This simplifies the problem. 3. subjects simply draw or otherwise specify different membership curves appropriate to a given problem. Subjective evaluation and elicitation As fuzzy sets are usually intended to model people's cognitive states. How are membership values determined? Determination methods break down broadly into the following categories: 1. Under more complex methods.0 which is one form of the fuzzy number 7. 2. Or they are given a more constrained set of possible curves from which they choose.0 7. most actual fuzzy control operations draw from a very small set of different curves. they can be determined from either simple or sophisticated elicitation procedures.5 + / \ | / \ | / \ 0.case is the triangular fuzzy number 1. each with its own mathematical and methodological strengths and weaknesses. Converted frequencies or probabilities Sometimes information taken in the form of frequency histograms or other probability curves are used as the basis to construct a membership function. it should always be remembered that membership functions are NOT .0 9. These subjects are typcially experts in the problem area.

In fact. fuzzy sets and probability exist as parts of a greater Generalized Information Theory which includes many formalisms for representing uncertainty (including random sets.).(necessarily) probabilities. probability intervals. from a mathematical perspective. that is that they must add together to one. fuzzy researchers have gone to great pains to distance themselves from probability. But this does not hold in general with membership grades. etc. so please refer to the following articles. fuzzy values are commonly misunderstood to be probabilities. Because of this. one can also talk about random fuzzy events and fuzzy random events. how does fuzzy theory differ from probability theory mathematically. general fuzzy measures. but almost none measure the membership grade directly. Physical measurement Many applications of fuzzy logic use physical measurement. how does it differ in interpretation and application. Semantically. See [10] for more information. many of them have lost track of another point. What is the relationship between fuzzy truth values and probabilities? This question has to be answered in two ways: first. and second. or the integral of their density curves must be one. And while membership grades can be determined with probability densities in mind (see [11]). This whole issue is beyond the scope of this FAQ. and then the individual membership grades of data are calculated from it . 4. interval analysis. they also generalize probability. At the mathematical level. there are other methods as well which have nothing to do with frequencies or probabilities. a membership function is provided by another method. Demster-Shafer evidence theory. the distinction between fuzzy logic and probability theory has to do with the difference between the notions of probability and a degree of membership. possibility theory. which is that the converse DOES hold: all probability distributions are fuzzy sets! As fuzzy sets and logic generalize Boolean sets and logic. But in so doing. Furthermore. or fuzzy logic is interpreted as some new way of handling probabilities. Probability statements are about the likelihoods of . A minimum requirement of probabilities is ADDITIVITY. Instead. But this is not the case. or the textbook by Klir and Folger (see [16]).

and instead you are trying to model the EXTENT to which an event occured. weights on transitions between states. The rules of possibility theory are similar to probability theory. all fuzzy numbers are possibility distributions. What is possibility theory? Possibility theory is a new form of information theory which is related to but independent of both fuzzy sets and probability theory. is to contact the organization that ran the conference or workshop you are interested in. Are there fuzzy state machines? Yes. the best idea mentioned so far is to contact the Institute for Scientific Information. or to Section III of the Dubois and Prade's 1980 textbook (see [16]). FSMs are obtained by assigning membership grades as weights to the states of a machine. and then a composition rule such as MAX/MIN or PLUS/TIMES (see [4]) to calculate new grades of future states. James C. but use either MAX/MIN or MAX/TIMES calculus. possibilistic NONSPECIFICITY is available as a measure of information similar to the stochastic ENTROPY. possibility theory can also be derived without reference to fuzzy sets. This issue is treated well in the swamp water example used by James Bezdek of the University of West Florida (Bezdek. of course. Also. and check with their .outcomes: an event either occurs or does not. How can I get a copy of the proceedings for ? This is rough sometimes. However. rather than the PLUS/TIMES calculus of probability theory. If they can't help you. because the PLUS/TIMES calculus does not validly generalize nondeterministic processes. Possibility theory has a methodological advantage over probability theory as a representation of nondeterminism in systems. But with fuzziness. "Fuzzy Models . Inc. The first thing to do. For example. Technically. one cannot say unequivocally whether an event occured or not. a possibility distribution is a normal fuzzy set (at least one membership grade equals 1). Refer to the following article. and you can bet on it. while MAX/MIN and MAX/TIMES do. (ISI).

.Index to Scientific and Technical Proceedings (ISTP volumes).

Sign up to vote on this title

UsefulNot usefulAre you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.