Fuzzy Stress Regulation Controller

A Mamdani fuzzy-logic model that converts two crisp inputs—stress level and arousal level—into a continuous feedback intensity from 0% to 100%. The repository was created for Practical Task 1 in the Intelligent Systems course and demonstrates model construction, inference, visualization, scenario testing, and membership-function parameter modification in Python.

Educational use only: this project is a fuzzy-control demonstration, not a medical device or diagnostic system.

Project Highlights

• Two input variables on a 0–10 scale: stress and arousal
• One output variable on a 0–100% scale: feedback intensity
• Three linguistic terms for each variable
• A complete nine-rule Mamdani rule base
• Centroid defuzzification through scikit-fuzzy
• Interactive and command-line execution modes
• Nine-scenario experiment with a heatmap and CSV export
• Baseline-versus-modified parameter experiment
• Automated tests for reference values, symmetry, monotonicity, and validation

Controller Design

Variable	Type	Range	Linguistic terms
Stress	Input	0–10	Low, Medium, High
Arousal	Input	0–10	Low, Medium, High
Feedback intensity	Output	0–100%	Gentle, Moderate, Strong

The baseline input membership functions are:

Term	Function	Parameters
Low	Trapezoidal	[0, 0, 2, 4]
Medium	Triangular	[2, 5, 8]
High	Trapezoidal	[6, 8, 10, 10]

The output membership functions are Gentle [0, 0, 15, 35], Moderate [20, 50, 80], and Strong [65, 85, 100, 100].

Rule Base

Stress \ Arousal	Low	Medium	High
Low	Gentle	Gentle	Moderate
Medium	Gentle	Moderate	Strong
High	Moderate	Strong	Strong

Example

For stress 8 and arousal 7, the baseline controller produces:

Feedback intensity: 84.86 %
Feedback category:  Strong
Suggested action:   Combined calming feedback and a rest reminder

Repository Structure

fuzzy-stress-controller/
├── .github/workflows/python-checks.yml
├── report/
│   ├── Fuzzy_Logic_Modeling_in_Python_Report_Mohammad_Azimi.docx
│   └── Fuzzy_Logic_Modeling_in_Python_Report_Mohammad_Azimi.pdf
├── results/
│   ├── membership_functions.png
│   ├── simulation_result.png
│   ├── scenario_comparison.csv
│   ├── scenario_comparison.png
│   ├── parameter_comparison.csv
│   ├── parameter_membership_comparison.png
│   └── parameter_output_comparison.png
├── tests/test_controller.py
├── main.py
├── scenario_experiments.py
├── parameter_comparison.py
└── requirements.txt

Installation on Windows

Open Command Prompt in the directory where you want to store the project:

git clone https://github.com/mohammad-azimi/fuzzy-stress-controller.git
cd fuzzy-stress-controller
py -3.12 -m venv .venv
.venv\Scripts\activate
python -m pip install --upgrade pip
python -m pip install -r requirements.txt

If the repository is already cloned, activate its existing environment:

cd G:\Projects\Uni-Projects\fuzzy-stress-controller
.venv\Scripts\activate

Running the Programs

1. Interactive controller

python main.py

Enter stress and arousal when prompted. The program prints the inference result, saves the plots in results/, and opens the final visualization.

The same example can be executed non-interactively:

python main.py --stress 8 --arousal 7

Use --no-show to generate files without opening a plot window:

python main.py --stress 8 --arousal 7 --no-show

2. Nine-scenario experiment

python scenario_experiments.py

This evaluates all representative Low, Medium, and High combinations and creates a CSV file and heatmap.

3. Parameter-modification experiment

python parameter_comparison.py

The modified input sets are shifted toward lower values:

Term	Baseline	Modified
Low	[0, 0, 2, 4]	[0, 0, 1.5, 3.5]
Medium	[2, 5, 8]	[1.5, 4.5, 7.5]
High	[6, 8, 10, 10]	[5, 7, 10, 10]

This makes the controller respond earlier in transition regions while leaving the rule base and output sets unchanged. The largest measured increase in the selected cases is 13.43 percentage points.

Automated Tests

Run the standard-library test suite with:

python -m unittest discover -s tests -v

The GitHub Actions workflow runs the same checks automatically after every push and pull request.

Report

• Project report (PDF)
• Project report (Word)

Academic Information

• Student: Mohammad Azimi
• Group: 5140901/61701
• Course: Intelligent Systems
• Instructor: Yuri Nurgalievich Kozhubaev
• Institution: Peter the Great St. Petersburg Polytechnic University
• Year: 2026

References

1. D. A. Novak, Yu. N. Kozhubaev, and E. N. Ovchinnikova, Modeling of Fuzzy Systems Controls in Python Programming, 2024.
2. L. A. Zadeh, “Fuzzy Sets,” Information and Control, 1965.
3. E. H. Mamdani and S. Assilian, “An Experiment in Linguistic Synthesis with a Fuzzy Logic Controller,” International Journal of Man-Machine Studies, 1975.
4. scikit-fuzzy documentation