FeelAnyForce: Estimating Contact Force Feedback from Tactile Sensation for Vision-Based Tactile Sensors
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913528904941568 |
|---|---|
| author | Shahidzadeh, Amir-Hossein Caddeo, Gabriele Alapati, Koushik Natale, Lorenzo Fermüller, Cornelia Aloimonos, Yiannis |
| author_facet | Shahidzadeh, Amir-Hossein Caddeo, Gabriele Alapati, Koushik Natale, Lorenzo Fermüller, Cornelia Aloimonos, Yiannis |
| contents | In this paper, we tackle the problem of estimating 3D contact forces using vision-based tactile sensors. In particular, our goal is to estimate contact forces over a large range (up to 15 N) on any objects while generalizing across different vision-based tactile sensors. Thus, we collected a dataset of over 200K indentations using a robotic arm that pressed various indenters onto a GelSight Mini sensor mounted on a force sensor and then used the data to train a multi-head transformer for force regression. Strong generalization is achieved via accurate data collection and multi-objective optimization that leverages depth contact images. Despite being trained only on primitive shapes and textures, the regressor achieves a mean absolute error of 4\% on a dataset of unseen real-world objects. We further evaluate our approach's generalization capability to other GelSight mini and DIGIT sensors, and propose a reproducible calibration procedure for adapting the pre-trained model to other vision-based sensors. Furthermore, the method was evaluated on real-world tasks, including weighing objects and controlling the deformation of delicate objects, which relies on accurate force feedback. Project webpage: http://prg.cs.umd.edu/FeelAnyForce |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_02048 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FeelAnyForce: Estimating Contact Force Feedback from Tactile Sensation for Vision-Based Tactile Sensors Shahidzadeh, Amir-Hossein Caddeo, Gabriele Alapati, Koushik Natale, Lorenzo Fermüller, Cornelia Aloimonos, Yiannis Robotics Computer Vision and Pattern Recognition In this paper, we tackle the problem of estimating 3D contact forces using vision-based tactile sensors. In particular, our goal is to estimate contact forces over a large range (up to 15 N) on any objects while generalizing across different vision-based tactile sensors. Thus, we collected a dataset of over 200K indentations using a robotic arm that pressed various indenters onto a GelSight Mini sensor mounted on a force sensor and then used the data to train a multi-head transformer for force regression. Strong generalization is achieved via accurate data collection and multi-objective optimization that leverages depth contact images. Despite being trained only on primitive shapes and textures, the regressor achieves a mean absolute error of 4\% on a dataset of unseen real-world objects. We further evaluate our approach's generalization capability to other GelSight mini and DIGIT sensors, and propose a reproducible calibration procedure for adapting the pre-trained model to other vision-based sensors. Furthermore, the method was evaluated on real-world tasks, including weighing objects and controlling the deformation of delicate objects, which relies on accurate force feedback. Project webpage: http://prg.cs.umd.edu/FeelAnyForce |
| title | FeelAnyForce: Estimating Contact Force Feedback from Tactile Sensation for Vision-Based Tactile Sensors |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.02048 |