Table of Contents
Fetching ...

Motion ReTouch: Motion Modification Using Four-Channel Bilateral Control

Koki Inami, Sho Sakaino, Toshiaki Tsuji

TL;DR

The paper addresses the challenge of achieving fast, complex manipulations with four-channel bilateral control by introducing Motion ReTouch, a post-editing approach that uses a tri-robot setup (leader, follower, editor) and multilateral control to retroactively modify motion data, including force information. By applying Motion ReTouch to data sped up by 3x in a test-tube transfer task, the authors demonstrate a complete recovery of task success and tangible changes in reaction forces while maintaining similar position trajectories. This work advances imitation-learning data editing, enabling more efficient generation of high-speed demonstrations and enhancing force-aware robotic manipulation. The practical impact lies in enabling faster and more reliable data-driven learning for contact-rich tasks without requiring new demonstrations from scratch.

Abstract

Recent research has demonstrated the usefulness of imitation learning in autonomous robot operation. In particular, teaching using four-channel bilateral control, which can obtain position and force information, has been proven effective. However, control performance that can easily execute high-speed, complex tasks in one go has not yet been achieved. We propose a method called Motion ReTouch, which retroactively modifies motion data obtained using four-channel bilateral control. The proposed method enables modification of not only position but also force information. This was achieved by the combination of multilateral control and motion-copying system. The proposed method was verified in experiments with a real robot, and the success rate of the test tube transfer task was improved, demonstrating the possibility of modification force information.

Motion ReTouch: Motion Modification Using Four-Channel Bilateral Control

TL;DR

The paper addresses the challenge of achieving fast, complex manipulations with four-channel bilateral control by introducing Motion ReTouch, a post-editing approach that uses a tri-robot setup (leader, follower, editor) and multilateral control to retroactively modify motion data, including force information. By applying Motion ReTouch to data sped up by 3x in a test-tube transfer task, the authors demonstrate a complete recovery of task success and tangible changes in reaction forces while maintaining similar position trajectories. This work advances imitation-learning data editing, enabling more efficient generation of high-speed demonstrations and enhancing force-aware robotic manipulation. The practical impact lies in enabling faster and more reliable data-driven learning for contact-rich tasks without requiring new demonstrations from scratch.

Abstract

Recent research has demonstrated the usefulness of imitation learning in autonomous robot operation. In particular, teaching using four-channel bilateral control, which can obtain position and force information, has been proven effective. However, control performance that can easily execute high-speed, complex tasks in one go has not yet been achieved. We propose a method called Motion ReTouch, which retroactively modifies motion data obtained using four-channel bilateral control. The proposed method enables modification of not only position but also force information. This was achieved by the combination of multilateral control and motion-copying system. The proposed method was verified in experiments with a real robot, and the success rate of the test tube transfer task was improved, demonstrating the possibility of modification force information.

Paper Structure

This paper contains 9 sections, 11 equations, 12 figures, 2 tables.

Figures (12)

  • Figure 1: 4-CH Bilateral Control
  • Figure 2: Multilateral Control
  • Figure 3: Motion Copying System
  • Figure 4: Motion ReTouch
  • Figure 6: Block Diagram of Motion ReTouch
  • ...and 7 more figures