Fuzzy Inference System Tuning

Tune membership functions and rules of fuzzy systems

You can tune the membership function parameters and rules of your fuzzy inference system using Global Optimization Toolbox tuning methods such as genetic algorithms and particle swarm optimization. For more information, see Tuning Fuzzy Inference Systems.

If your system is a single-output type-1 Sugeno FIS, you can tune its membership function parameters using neuro-adaptive learning methods. This tuning method does not require Global Optimization Toolbox software. For more information, see Neuro-Adaptive Learning and ANFIS.

Apps

Fuzzy Logic Designer

Design, test, and tune fuzzy inference systems

Functions

expand all

Tune Fuzzy Systems

`tunefis`	Tune fuzzy inference system or tree of fuzzy inference systems
`tunefisOptions`	Option set for `tunefis` function
`getTunableSettings`	Obtain tunable settings from fuzzy inference system
`setTunable`	Set specified parameter settings as tunable or nontunable
`getTunableValues`	Obtain values of tunable parameters from fuzzy inference system
`setTunableValues`	Specify tunable parameter values of a fuzzy inference system

Train ANFIS

`anfis`	Tune Sugeno-type fuzzy inference system using training data
`anfisOptions`	Option set for `anfis` function

Objects

expand all

Tunable Settings

`RuleSettings`	Tunable parameter settings of fuzzy rules
`VariableSettings`	Tunable parameter settings of fuzzy variables
`MembershipFunctionSettings`	Tunable parameter settings for fuzzy membership functions
`MembershipFunctionSettingsType2`	Tunable parameter settings for type-2 fuzzy membership functions
`ClauseParameters`	Parameter settings for rule clauses
`NumericParameters`	Tunable numeric parameter settings of membership functions

Topics

Tune Fuzzy Systems

Tuning Fuzzy Inference Systems
Tune fuzzy membership function parameters and learn new fuzzy rules.
Tune Fuzzy Inference System Using Fuzzy Logic Designer

Interactively learn rules and tune membership function parameters of a fuzzy inference system.
- Configure Tuning Options in Fuzzy Logic Designer
- Select Rules and Parameters to Tune in Fuzzy Logic Designer
Tune Fuzzy Inference System at the Command Line

Programmatically learn rules and tune membership function parameters of a fuzzy inference system.
- Select FIS Rules and Parameters to Tune at the Command Line
Tune Fuzzy Trees
You can tune the learn rules and tune membership function parameters for FISs within a fuzzy tree.
Customize FIS Tuning Process
You can customize the FIS tuning process by specifying either a custom cost function or a custom optimization method.
Optimize FIS Parameters with K-Fold Cross-Validation
To prevent overfitting during FIS parameter optimization, you can stop the tuning process early based on an unbiased evaluation of the model using validation data.
Tune FIS Tree for Gas Mileage Prediction
Tune the rules and membership function parameters for a tree of interconnected Sugeno fuzzy systems.
Predict Chaotic Time Series Using Type-2 FIS
Tune the rules and membership function parameters for a FIS with type-2 membership functions.
Tune Fuzzy Robot Obstacle Avoidance System Using Custom Cost Function
When you do not have training data, you can tune your fuzzy system using a custom cost function that simulates the FIS operation.

Train ANFIS Systems

Neuro-Adaptive Learning and ANFIS
You can tune Sugeno fuzzy inference systems using neuro-adaptive learning techniques similar to those used for training neural networks.
Train Adaptive Neuro-Fuzzy Inference Systems
Interactively create, train, and test neuro-fuzzy systems using the Fuzzy Logic Designer app.
Predict Chaotic Time-Series Using ANFIS
Train a neuro-fuzzy system for time-series prediction using the anfis command.
Adaptive Noise Cancellation Using ANFIS
Perform adaptive nonlinear noise cancellation using the anfis and genfis commands.
Model Suburban Commuting Using Subtractive Clustering and ANFIS
Generate a fuzzy inference system from data using subtractive clustering.
Gas Mileage Prediction
Predict fuel consumption for automobiles using an adaptive neuro-fuzzy inference system and previously recorded observations.
Nonlinear System Identification
You can model nonlinear dynamic system behavior using adaptive neuro-fuzzy systems.

Featured Examples

Classify Pixels Using Fuzzy Systems

Tune separate fuzzy inference systems to classify pixels based on color and texture, and combine these systems into a fuzzy tree for image segmentation.

Open Live Script

Autonomous Parking Using Fuzzy Inference System

Build and tune a FIS tree for autonomous parking.

Open Live Script

Design Controller for Artificial Pancreas Using Fuzzy Logic

Design and tune a FIS tree to control insulin infusion for type-1 diabetes.

Open Live Script

Explain Black-Box Model Using Fuzzy Support System

Explain the behavior of a trained deep neural network by tuning a fuzzy inference system when the original training data is available.

Open Live Script

Explainable Fuzzy Support System for Black-Box Model of Robot Obstacle Avoidance

Explain the behavior of a trained reinforcement learning agent by tuning a fuzzy inference system when the original training data is not available.

Open Live Script

Modeling Inverse Kinematics in a Robotic Arm

Determine joint angles required to place the tip of a robotic arm in a desired location using a neuro-fuzzy model.

Open Live Script