Fuzzy inference is the process of formulating input/output mappings using fuzzy logic. Fuzzy Logic Toolbox™ software provides command-line functions and an app for creating Mamdani and Sugeno fuzzy systems. For more information on fuzzy logic, see What Is Fuzzy Logic?. For more information on fuzzy inference, see Fuzzy Inference Process.
|Fuzzy Logic Designer||Design and test fuzzy inference systems|
|Create new fuzzy inference system|
|Generate fuzzy inference system structure from data|
|Option set for genfis command|
|Add variable to fuzzy inference system|
|Remove variables from fuzzy inference system|
|Transform Mamdani Fuzzy Inference System into Sugeno Fuzzy Inference System|
|Get fuzzy system properties|
|Set fuzzy system properties|
|Gaussian curve membership function|
|Generalized bell-shaped membership function|
|Triangular-shaped membership function|
|Difference between two sigmoidal membership functions|
|Gaussian combination membership function|
|Π-shaped membership function|
|Product of two sigmoidal membership functions|
|Sigmoidal membership function|
|S-shaped membership function|
|Trapezoidal-shaped membership function|
|Z-shaped membership function|
|Plot all membership functions for given variable|
|Generic membership function evaluation|
|Translate parameters between membership functions|
|Defuzzify membership function|
|Perform fuzzy arithmetic|
Fuzzy inference maps an input space to an output space using a series of fuzzy if-then rules.
In Mamdani systems, the output of each if-then rule is a fuzzy set. To determine a crisp output value, these fuzzy sets are aggregated and defuzzified.
In Sugeno systems, the output of each if-then rule is either constant or a linear function of the input variables. The final output value is the weighted average of all rule outputs.
Both Mamdani and Sugeno systems have several advantages depending on your specific application.
Interactively construct a Mamdani fuzzy inference system using the Fuzzy Logic Designer app.
Construct a Mamdani fuzzy inference system at the MATLAB® command line.
You can replace the built-in membership functions and fuzzy inference functions with your own custom functions.