Learning Bayes Filter Models for Tactile Localization

TL;DR

This paper introduces learnable Bayes filter models that utilize tactile feedback and visual maps to localize robotic grippers, especially useful for low-precision or poorly calibrated manipulators, with successful simulation and real-world results.

Contribution

The paper presents a novel tactile-based localization method using learnable Bayes filters conditioned on visual maps, trained in simulation and transferred to real robots.

Findings

Effective localization in simulation and real-world scenarios.

Generalizes across different object sizes, shapes, and configurations.

Outperforms baseline methods in tactile localization tasks.

Abstract

Localizing and tracking the pose of robotic grippers are necessary skills for manipulation tasks. However, the manipulators with imprecise kinematic models (e.g. low-cost arms) or manipulators with unknown world coordinates (e.g. poor camera-arm calibration) cannot locate the gripper with respect to the world. In these circumstances, we can leverage tactile feedback between the gripper and the environment. In this paper, we present learnable Bayes filter models that can localize robotic grippers using tactile feedback. We propose a novel observation model that conditions the tactile feedback on visual maps of the environment along with a motion model to recursively estimate the gripper's location. Our models are trained in simulation with self-supervision and transferred to the real world. Our method is evaluated on a tabletop localization task in which the gripper interacts with objects. We report results in simulation and on a real robot, generalizing over different sizes, shapes, and configurations of the objects.