| Abstract Scope |
The sense of touch gathers information, such as temperature and smoothness, through direct contact and makes tasks such as manipulating objects possible. Tactile sensors replicate one or more dimensions of the sense and have potential applications in robotics and quality control. In this research, I used a simple tactile sensor and machine learning algorithms to analyze materials. Two datasets are created, one with common materials such as felt, cotton, and wood, and another with wrinkled and smooth paper to simulate defects. Two Machine Learning models, Support Vector Machine and Convolutional Neural Networks, evaluated the datasets. After optimization, both models reached accuracies around 90% for both datasets, demonstrating that applying machine learning to material classification and defect detection is a viable research direction. |