Inhaltspezifische Aktionen

Active Sampling for Hardness Classification with Vision-Based Tactile Sensors

Junyi Chen, Alap Kshirsagar, Frederik Heller, Mario Gómez Andreu, Boris Belousov, Tim Schneider, Lisa P. Y. Lin, Katja Doerschner, Knut Drewing & Jan Peters

Hardness is a key tactile property perceived by humans and robots. In this work, we investigate informationtheoretic active sampling for efficient hardness classificationusing vision-based tactile sensors. We assess three probabilisticclassifiers and two uncertainty-based sampling strategies on arobotic setup and a human-collected dataset. Results show thatuncertainty-driven sampling outperforms random sampling in accuracy and stability. While human participants achieve 48.00% accuracy, our best method reaches 88.78% on the same objects, highlighting the effectiveness of vision-based tactilesensors for hardness classification.
Chen, J., Kshirsagar, A., Heller, F., Andreu, M. G., Belousov, B., Schneider, T., ... & Peters, J. Active Sampling for Hardness Classification with Vision-Based Tactile Sensors.