Tactile perception for texture and object characterization
Robots capable of classifying objects and textures will provide early insights when interacting with their environment. There is a large area of application of such robots, from garbage selection to domestic tasks, capable of characterization.
During this project, I and some master students performed experiments using two types of multi-modal tactile sensors 1.
Using inertial measurements and pressure, information from tactile sensor trained machine learning models to classify textures and sets of objects.
de Oliveira, T. E. A., Prado da Fonseca, V., Huluta, E., Rosa, P. F. F., & Petriu, E. M. (2015). Data-driven analysis of kinaesthetic and tactile information for shape classification. 2015 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA), 1–5. https://doi.org/10.1109/CIVEMSA.2015.7158615 ↩