Precise Object Placement Using Force-Torque Feedback

TL;DR

This paper presents a force-torque feedback method enabling robots to accurately and robustly place objects in stable positions, outperforming vision-based approaches especially in noisy or adversarial conditions.

Contribution

It introduces a novel force-torque sensing approach for precise object placement that can recover from errors and sensor noise, improving robustness over prior vision-based methods.

Findings

100% success in basic stacking tasks

17% success in complex adjustment scenarios

Robust placement in adversarial environments

Abstract

Precise object manipulation and placement is a common problem for household robots, surgery robots, and robots working on in-situ construction. Prior work using computer vision, depth sensors, and reinforcement learning lacks the ability to reactively recover from planning errors, execution errors, or sensor noise. This work introduces a method that uses force-torque sensing to robustly place objects in stable poses, even in adversarial environments. On 46 trials, our method finds success rates of 100% for basic stacking, and 17% for cases requiring adjustment.