Deploy Fuzzy Inference Systems

You can deploy your fuzzy inference system (FIS) by:

Generating code in MATLAB^® or Simulink^®
Packaging and deploying a standalone application in MATLAB or Simulink

All fuzzy inference system options, including custom inference functions, support code generation and deployment.

Generate Code in MATLAB

You can generate code for evaluating the following types of fuzzy inference systems using MATLAB Coder™.

Type-1 fuzzy inference systems (mamfis, sugfis)
Type-2 fuzzy inference systems (mamfistype2, sugfistype2)
FIS trees (fistree) (since R2022b)

Code generation in MATLAB does not directly support fuzzy inference system objects. Instead, you must first convert your fuzzy system into a homogeneous structure using getFISCodeGenerationData, and then pass the resulting structure to evalfis.

Code generation for fuzzy systems in MATLAB supports double-precision and single-precision data.

For more information, see Generate Code for Fuzzy System Using MATLAB Coder.

Note

Code generation does not support the construction of fuzzy systems at the command line.

Generate Code in Simulink

You can generate code for evaluating fuzzy inference systems implemented in Simulink using Simulink Coder or Simulink PLC Coder™.

Simulink code generation supports the following blocks:

Fuzzy Logic Controller
FIS Tree (since R2024a)

Code generation for fuzzy systems in Simulink supports double-precision, single-precision, and fixed-point data.

For more information, see Generate Code for Fuzzy System Using Simulink Coder and Generate Structured Text for Fuzzy System Using Simulink PLC Coder.

Deploy Fuzzy Systems

You can package and deploy standalone applications that include fuzzy systems using either MATLAB Compiler™ or Simulink Compiler.

You can compile:

MATLAB code that uses most Fuzzy Logic Toolbox™ command-line functionality. Compiling code that uses graphical functions (plotfis, plotmf, gensurf) is not supported. Deploying the Fuzzy Logic Designer app is not supported. For more information on compiling MATLAB code, see Create Standalone Application from MATLAB Function (MATLAB Compiler).
Simulink models that use the Fuzzy Logic Controller block. For more information on compiling Simulink models, see Create and Deploy a Script with Simulink Compiler (Simulink Compiler).