Optimal Deep Learning for Robot Touch

TL;DR

This paper demonstrates how deep learning can be used to accurately estimate the relative pose of objects through tactile sensors, enhancing robotic manipulation capabilities.

Contribution

It introduces a method for training deep neural networks to estimate object pose from tactile data, incorporating unlabelled perturbations and Bayesian hyperparameter optimization.

Findings

Deep learning models can accurately estimate object pose from tactile data.

Inclusion of unlabelled perturbations improves model robustness.

Bayesian optimization enhances model accuracy and hyperparameter tuning.

Abstract

This article illustrates the application of deep learning to robot touch by considering a basic yet fundamental capability: estimating the relative pose of part of an object in contact with a tactile sensor. We begin by surveying deep learning applied to tactile robotics, focussing on optical tactile sensors, which help bridge from deep learning for vision to touch. We then show how deep learning can be used to train accurate pose models of 3D surfaces and edges that are insensitive to nuisance variables such as motion-dependent shear. This involves including representative motions as unlabelled perturbations of the training data and using Bayesian optimization of the network and training hyperparameters to find the most accurate models. Accurate estimation of pose from touch will enable robots to safely and precisely control their physical interactions, underlying a wide range of object exploration and manipulation tasks.