# Robust Learning of Tactile Force Estimation through Robot Interaction

arXiv: https://arxiv.org/abs/1810.06187 · HTML: https://mlsanity.com/paper/1810.06187/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Oct 2018.

## Built on by (most cited first, 7 of 7)

- [Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AI](https://mlsanity.com/paper/2407.06886/) (2024)
- [ReSkin: versatile, replaceable, lasting tactile skins](https://mlsanity.com/paper/2111.00071/) (2021)
- [VTAM: Video-Tactile-Action Models for Complex Physical Interaction Beyond VLAs](https://mlsanity.com/paper/2603.23481/) (2026)
- [Hearing Touch: Audio-Visual Pretraining for Contact-Rich Manipulation](https://mlsanity.com/paper/2405.08576/) (2024)
- [FeelAnyForce: Estimating Contact Force Feedback from Tactile Sensation for Vision-Based Tactile Sensors](https://mlsanity.com/paper/2410.02048/) (2024)
- [Action Conditioned Tactile Prediction: case study on slip prediction](https://mlsanity.com/paper/2205.09430/) (2022)
- [Planning Visual-Tactile Precision Grasps via Complementary Use of Vision and Touch](https://mlsanity.com/paper/2212.08604/) (2022)

## Abstract (arXiv)

Current methods for estimating force from tactile sensor signals are either inaccurate analytic models or task-specific learned models. In this paper, we explore learning a robust model that maps tactile sensor signals to force. We specifically explore learning a mapping for the SynTouch BioTac sensor via neural networks. We propose a voxelized input feature layer for spatial signals and leverage information about the sensor surface to regularize the loss function. To learn a robust tactile force model that transfers across tasks, we generate ground truth data from three different sources: (1) the BioTac rigidly mounted to a force torque~(FT) sensor, (2) a robot interacting with a ball rigidly attached to the same FT sensor, and (3) through force inference on a planar pushing task by formalizing the mechanics as a system of particles and optimizing over the object motion. A total of 140k samples were collected from the three sources. We achieve a median angular accuracy of 3.5 degrees in predicting force direction (66% improvement over the current state of the art) and a median magnitude accuracy of 0.06 N (93% improvement) on a test dataset. Additionally, we evaluate the learned force model in a force feedback grasp controller performing object lifting and gentle placement. Our results can be found on https://sites.google.com/view/tactile-force.