Real Garment Benchmark (RGBench): A Comprehensive Benchmark for Robotic Garment Manipulation featuring a High-Fidelity Scalable Simulator

TL;DR

RGBench introduces a comprehensive benchmark with a large dataset and a high-fidelity simulator to advance robotic manipulation of garments, addressing the lack of realistic deformable object simulation.

Contribution

It provides a new extensive dataset, a high-performance simulator, and evaluation protocols for deformable object manipulation, significantly improving simulation accuracy and speed.

Findings

Simulator reduces error by 20% compared to existing options.

Simulator runs three times faster than current cloth simulators.

RGBench will be publicly available to support future research.

Abstract

While there has been significant progress to use simulated data to learn robotic manipulation of rigid objects, applying its success to deformable objects has been hindered by the lack of both deformable object models and realistic non-rigid body simulators. In this paper, we present Real Garment Benchmark (RGBench), a comprehensive benchmark for robotic manipulation of garments. It features a diverse set of over 6000 garment mesh models, a new high-performance simulator, and a comprehensive protocol to evaluate garment simulation quality with carefully measured real garment dynamics. Our experiments demonstrate that our simulator outperforms currently available cloth simulators by a large margin, reducing simulation error by 20% while maintaining a speed of 3 times faster. We will publicly release RGBench to accelerate future research in robotic garment manipulation. Website: https://rgbench.github.io/