Diffusion-EDFs:
Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation
CVPR 2024 (Highlight)
Abstract
Diffusion generative modeling has become a promising approach for learning robotic manipulation tasks from stochastic human demonstrations. In this paper, we present Diffusion-EDFs, a novel SE(3)-equivariant diffusion-based approach for visual robotic manipulation tasks. We show that our proposed method achieves remarkable data efficiency, requiring only 5 to 10 human demonstrations for effective end-to-end training in less than an hour. Furthermore, our benchmark experiments demonstrate that our approach has superior generalizability and robustness compared to state-of-the-art methods. Lastly, we validate our methods with real hardware experiments.
Gallery
Technical Demonstration @ Yonsei University M.E. Grad. Students Conference
(Best Technical Demonstration Award)
Real robot experiments
Mug-on-a-hanger
Bottles-on-a-shelf
Bowls-on-dishes
Citation
Please consider citing our paper if you find it helpful.
@article{ryu2023diffusion,
title={Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation},
author={Ryu, Hyunwoo and Kim, Jiwoo and Chang, Junwoo and Ahn, Hyun Seok and Seo, Joohwan and Kim, Taehan and Choi, Jongeun and Horowitz, Roberto},
journal={arXiv preprint arXiv:2309.02685},
year={2023}
}