Fuzzy Logic Toolbox

Design and simulate fuzzy logic systems

Fuzzy Logic Toolbox™ provides MATLAB^® functions, apps, and a Simulink^® block for analyzing, designing, and simulating fuzzy logic systems. The product lets you specify and configure inputs, outputs, membership functions, and rules of type-1 and type-2 fuzzy inference systems.

The toolbox lets you automatically tune membership functions and rules of a fuzzy inference system from data. You can evaluate the designed fuzzy logic systems in MATLAB and Simulink. Additionally, you can use the fuzzy inference system as a support system to explain artificial intelligence (AI)-based black-box models. You can generate standalone executables or C/C++ code and IEC 61131-3 Structured Text to evaluate and implement fuzzy logic systems.

Fuzzy Logic Designer

Use the Fuzzy Logic Designer app or command-line functions to interactively design and simulate fuzzy inference systems. Define input and output variables and membership functions. Specify fuzzy if-then rules. Evaluate your fuzzy inference system across multiple input combinations.

Documentation | Examples

Fuzzy Inference Systems (FIS)

Implement Mamdani and Sugeno fuzzy inference systems. Convert from a Mamdani system to a Sugeno system or vice versa, to create and compare multiple designs. Additionally, implement complex fuzzy inference systems as a collection of smaller interconnected fuzzy systems using fuzzy trees.

Documentation | Examples

A screenshot of the Fuzzy Logic Designer app showing a type-2 membership function.

Type-2 Fuzzy Logic

Create and evaluate interval type-2 fuzzy inference systems with additional membership function uncertainty. Create type-2 Mamdani and Sugeno fuzzy inference systems using the Fuzzy Logic Designer app or using toolbox functions.

Documentation | Examples

A plot of a Mackey-Glass (MG) time series and a predicted time series using an adaptive neuro-fuzzy inference system (ANFIS).

Fuzzy Inference System Tuning

Tune membership function parameters and rules of a single fuzzy inference system or of a fuzzy tree using genetic algorithms, particle swarm optimization, and other Global Optimization Toolbox tuning methods. Train Sugeno fuzzy inference systems using neuro-adaptive learning techniques similar to those used for training neural networks.

Documentation | Examples

Cluster Centers of Iris Data Identified Using a Fuzzy C-Means Algorithm

Fuzzy Clustering

Find clusters in input/output data using fuzzy c-means or subtractive clustering. Use the resulting cluster information to generate a Sugeno-type fuzzy inference system that models the input/output data behavior.

Documentation | Examples

A Simulink model containing a Fuzzy Logic Controller Block to implement a fuzzy inference system.

Fuzzy Logic in Simulink

Evaluate and test the performance of your fuzzy inference system in Simulink using the Fuzzy Logic Controller block. Implement your fuzzy inference system as part of a larger system model in Simulink for system-level simulation and code generation.

Documentation | Examples

A script showing generated code for evaluating a fuzzy inference system.

Fuzzy Logic Deployment

Implement your fuzzy inference system in Simulink and generate C/C++ code or IEC61131-3 Structured Text using Simulink Coder™ or Simulink PLC Coder™, respectively. Use MATLAB Coder™ to generate C/C++ code from fuzzy inference systems implemented in MATLAB. Alternatively, compile your fuzzy inference system as a standalone application using MATLAB Compiler™.

Documentation | Examples

A diagram showing the flow of control between a black-box model and a fuzzy system for run-time explanation of black-box predictions.

Fuzzy Logic for Explainable AI

Use fuzzy inference systems as support systems to explain the input-output relationships modeled by an AI-based black-box system. Interpret the decision-making process of a black-box model using the explainable rule base of your fuzzy inference system.

Documentation | Examples