Here is a selection of projects I’ve worked on : |
Modeling of an intelligent braking system using Simulink, integrating Anti-lock Braking System (ABS) and Electronic Brakeforce Distribution (EBD).
Concepts: Vehicle dynamics, safety systems
Language: MATLAB / Simulink
Download Report
Recommendation system using KNN on the Book-Crossings dataset, analyzing user-book similarity and preferences.
Concepts: Machine Learning, K-Nearest Neighbors (KNN)
Language: Python (Pandas, scikit-learn)
View on GitHub
Real-time navigation, SLAM, and obstacle avoidance using ROS, with sensor integration and RViz visualization.
Concepts: Mobile robotics, real-time control
Language: Python / C++ / ROS1
Download Report
Classifier with grid search on weather data, evaluated by metrics.
Concepts: Classification, cross-validation
Language: Python
Logistic Regression pipeline to predict passenger survival using engineered features. Coefficient analysis, model tuning, and performance evaluation with accuracy and interpretability metrics.
Concepts: Logistic regression, feature coefficients, supervised learning
Language: Python
View on GitHub
AI-powered bot using multi-level Markov models and adaptive strategy with fallback on frequency analysis.
Concepts: Markov chains, adaptive learning
Language: Python
Interactive interface for vehicle stability (MF400H quad & Mercedes S-Class) with real-time dynamic modeling and safety indicators.
Concepts: Dynamic modeling, simulation
Language: MATLAB
.webp)
Python/OpenCV pipeline for object detection on grayscale images using adaptive thresholding, contour extraction, and morphology.
Concepts: Computer vision, image segmentation
Language: Python / OpenCV
CNN-based image classifier built from scratch with data preprocessing, model training and evaluation using TensorFlow.
Concepts: Image classification, CNN
Language: Python / TensorFlow
Keras regression model predicting medical expenses using demographic and lifestyle features with one-hot encoding and early stopping.
Concepts: Regression, data preprocessing, Keras
Language: Python
RNN-based spam detector trained on public text datasets using TensorFlow and Keras.
Concepts: Natural Language Processing (NLP), recurrent neural networks
Language: Python (TensorFlow / Keras)
C/C# suite for UART-based robot: real-time commands, telemetry monitoring, UI design and MCU testing.
Concepts: Embedded systems, UART communication
Language: C / C#
Satellite and bathymetry fusion for seagrass mapping using QGIS and supervised classification techniques.
Concepts: Supervised classification, GIS analysis
Language: QGIS / Python
Unsupervised clustering comparison using DBSCAN and HDBSCAN, with evaluation of density-based methods and parameter sensitivity.
Concepts: Clustering, unsupervised learning, density estimation
Language: Python (scikit-learn, hdbscan, matplotlib)
Supervised classification using decision trees, including feature importance, model visualization and accuracy evaluation.
Concepts: Decision trees, classification, model interpretability
Language: Python (scikit-learn, matplotlib)
Comparison of classification performance using precision, recall, F1-score, ROC curves and confusion matrices on supervised models.
Concepts: Classification metrics, model evaluation, ROC analysis
Language: Python (scikit-learn, matplotlib, seaborn)
Analysis of K-Means clustering performance using inertia, silhouette score and visual inspection across different k values.
Concepts: Unsupervised learning, clustering evaluation, K-Means
Language: Python (scikit-learn, matplotlib, seaborn)
Unsupervised clustering of customer data using K-Means to identify distinct market segments based on purchasing behavior.
Concepts: Customer segmentation, clustering, K-Means
Language: Python (scikit-learn, pandas, seaborn)
Analysis of Random Forest performance using feature importance, classification metrics, and comparison with simpler models.
Concepts: Ensemble methods, feature importance, supervised learning
Language: Python (scikit-learn, matplotlib, seaborn)