Robust Switched Control with Guaranteed Cost of a Furuta Pendulum Prototype using T-S Fuzzy mode: Data
Description
This repository contains the complete experimental datasets and MATLAB codes for reproducing Figures 2 and 3 from the paper **"Switched Fuzzy Control with Guaranteed Cost of a Furuta Pendulum Prototype"**, accepted for presentation at the IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) 2026 / WCCI 2026. The repository enables full reproducibility of the comparative analysis, controller design, and experimental validation results presented in the publication. ## Video Demonstration A video of the Furuta pendulum experiment with the implemented switched fuzzy controller is available on YouTube: https://www.youtube.com/watch?v=i5m7Hab6jYk ## Repository Structure ### 📁 Data Files - **Furuta_Pendulum_longer_rod.mat**: Experimental dataset for the longer pendulum configuration (states, control signals, disturbances) - **Furuta_Pendulum_short_rod.mat**: Experimental dataset for the shorter pendulum configuration - **switched_controller_design_beta_0.9.mat**: Designed switched controller with decay rate β=0.9 (gain matrices and parameters) - **Data_switched_gain_mu_M_*.mat**: LMI-based controller design results for switched control with parameters μ_M = 2000, 1000, 500, 100 - **Data_single_gain_mu_M_*.mat**: LMI-based controller design results for single gain control with parameters μ_M = 2000, 1000, 500, 100 ### 📁 MATLAB Code Files - **MAIN_furuta_paper_data_pipeline.m**: Main script to run the entire data generation pipeline - **generate_TS_fuzzy_model_furuta_pendulum.m**: Generates Takagi-Sugeno fuzzy model (polytopic representation) - **generate_TS_hybrid_uncertainty_model.m**: Creates hybrid uncertainty models with norm-bounded uncertainties - **generate_figure2_switched_vs_single_gain_data.m**: Computes LMI solutions for switched vs. single gain control comparison (Figure 2 data) - **Figure_2_guaranteed_cost_comparison.m**: MATLAB script to generate Figure 2 from saved data - **Figure_3_experimental_results_Furuta_prototype.m**: MATLAB script to generate Figure 3 showing experimental results for both pendulum lengths - **design_switched_controller.m**: Example script for designing a single switched controller with fixed β ### 📁 Generated Figures (Publication Ready) - **Figure_2_switched_vs_single_gain_comparison.eps/.jpg**: Figure 2 in vector (EPS) and raster (JPEG) formats - **Figure_3_Furuta_experimental_results.eps/.jpg**: Figure 3 in vector (EPS) and raster (JPEG) formats ## Key Features - **Complete Data Availability**: All experimental measurements and controller design results - **Full Reproducibility**: Includes all MATLAB codes for TS fuzzy modeling, hybrid uncertainty, LMI-based design, and figure generation - **MATLAB Compatibility**: Tested with MATLAB R2020a and later versions (requires YALMIP and an LMI solver) ## Citation If you use these datasets, codes, or figures in your research, please cite the corresponding paper (accepted for FUZZ-IEEE 2026).
Files
Steps to reproduce
## Quick Start 1. Clone this repository and open MATLAB in the repository directory 2. Run `MAIN_furuta_paper_data_pipeline.m` to generate all models, design data, and figures from scratch 3. Alternatively, run `Figure_2_guaranteed_cost_comparison.m` or `Figure_3_experimental_results_Furuta_prototype.m` to regenerate individual figures using precomputed data 4. Generated figures will be saved as both EPS (publication quality) and JPEG (presentation quality) files
Institutions
- Universidade Estadual Paulista Julio de Mesquita Filho Faculdade de Engenharia Campus de Ilha SolteiraSP, Ilha Solteira