Robust Learning of Tactile Force Estimation through Robot Interaction

TL;DR

This paper presents a neural network-based tactile force estimation model that is robust across tasks, leveraging spatial features and multiple data sources, significantly improving accuracy over previous methods.

Contribution

The authors introduce a novel voxelized input layer and a multi-source data collection approach to enhance tactile force estimation robustness and transferability.

Findings

Achieved 3.5° median angular force direction accuracy, 66% better than prior art.

Achieved 0.06 N median force magnitude accuracy, 93% improvement.

Successfully applied the model in a force feedback grasping task.

Abstract

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.