Table of Contents
Fetching ...

DTactive: A Vision-Based Tactile Sensor with Active Surface

Jikai Xu, Lei Wu, Changyi Lin, Ding Zhao, Huazhe Xu

TL;DR

This work presents DTactive, a novel vision-based tactile sensor with active surfaces that inherits and modifies the tactile 3D shape reconstruction method of DTact while integrating a mechanical transmission mechanism that facilitates the mobility of its surface.

Abstract

The development of vision-based tactile sensors has significantly enhanced robots' perception and manipulation capabilities, especially for tasks requiring contact-rich interactions with objects. In this work, we present DTactive, a novel vision-based tactile sensor with active surfaces. DTactive inherits and modifies the tactile 3D shape reconstruction method of DTact while integrating a mechanical transmission mechanism that facilitates the mobility of its surface. Thanks to this design, the sensor is capable of simultaneously performing tactile perception and in-hand manipulation with surface movement. Leveraging the high-resolution tactile images from the sensor and the magnetic encoder data from the transmission mechanism, we propose a learning-based method to enable precise angular trajectory control during in-hand manipulation. In our experiments, we successfully achieved accurate rolling manipulation within the range of [ -180°,180° ] on various objects, with the root mean square error between the desired and actual angular trajectories being less than 12° on nine trained objects and less than 19° on three novel objects. The results demonstrate the potential of DTactive for in-hand object manipulation in terms of effectiveness, robustness and precision.

DTactive: A Vision-Based Tactile Sensor with Active Surface

TL;DR

This work presents DTactive, a novel vision-based tactile sensor with active surfaces that inherits and modifies the tactile 3D shape reconstruction method of DTact while integrating a mechanical transmission mechanism that facilitates the mobility of its surface.

Abstract

The development of vision-based tactile sensors has significantly enhanced robots' perception and manipulation capabilities, especially for tasks requiring contact-rich interactions with objects. In this work, we present DTactive, a novel vision-based tactile sensor with active surfaces. DTactive inherits and modifies the tactile 3D shape reconstruction method of DTact while integrating a mechanical transmission mechanism that facilitates the mobility of its surface. Thanks to this design, the sensor is capable of simultaneously performing tactile perception and in-hand manipulation with surface movement. Leveraging the high-resolution tactile images from the sensor and the magnetic encoder data from the transmission mechanism, we propose a learning-based method to enable precise angular trajectory control during in-hand manipulation. In our experiments, we successfully achieved accurate rolling manipulation within the range of [ -180°,180° ] on various objects, with the root mean square error between the desired and actual angular trajectories being less than 12° on nine trained objects and less than 19° on three novel objects. The results demonstrate the potential of DTactive for in-hand object manipulation in terms of effectiveness, robustness and precision.

Paper Structure

This paper contains 14 sections, 12 equations, 9 figures, 1 table.

Figures (9)

  • Figure 1: A two-finger gripper is equipped with DTactive at its ends, using the tactile feedback and active surface of the sensors to perform bottle cap flipping manipulation. The numbered images represent the sequential steps of the operation.
  • Figure 2: Design of DTactive. (a) Components of DTactive are shown in the sectional view. (b) Front view of DTactive, showing the external transmission mechanism. (c) Rear view of DTactive.
  • Figure 3: Fabrication of the essential components of DTactive. (a) Fabrication of the contact module. (b) Fabrication of the supporting resin layer.
  • Figure 4: Tactile 3D reconstruction. (a) The tactile image without lubricating oil. (b) The tactile image with lubricating oil applied. (c) The tactile image when a badge with printed letters is pressed onto the sensor. (d) Depth map corresponding to the captured image in (c).
  • Figure 5: (a) Comparison of the minimum object radius that can be grasped by DTactive and a roller grasper. (b) Force analysis of an object under external moment (M) during grasping. (c) Force components involved in lifting an object.
  • ...and 4 more figures