Diffusion Policy: Visuomotor Policy Learning via Action Diffusion

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chi, Cheng, Xu, Zhenjia, Feng, Siyuan, Cousineau, Eric, Du, Yilun, Burchfiel, Benjamin, Tedrake, Russ, Song, Shuran
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913264229679104
author Chi, Cheng
Xu, Zhenjia
Feng, Siyuan
Cousineau, Eric
Du, Yilun
Burchfiel, Benjamin
Tedrake, Russ
Song, Shuran
author_facet Chi, Cheng
Xu, Zhenjia
Feng, Siyuan
Cousineau, Eric
Du, Yilun
Burchfiel, Benjamin
Tedrake, Russ
Song, Shuran
contents This paper introduces Diffusion Policy, a new way of generating robot behavior by representing a robot's visuomotor policy as a conditional denoising diffusion process. We benchmark Diffusion Policy across 12 different tasks from 4 different robot manipulation benchmarks and find that it consistently outperforms existing state-of-the-art robot learning methods with an average improvement of 46.9%. Diffusion Policy learns the gradient of the action-distribution score function and iteratively optimizes with respect to this gradient field during inference via a series of stochastic Langevin dynamics steps. We find that the diffusion formulation yields powerful advantages when used for robot policies, including gracefully handling multimodal action distributions, being suitable for high-dimensional action spaces, and exhibiting impressive training stability. To fully unlock the potential of diffusion models for visuomotor policy learning on physical robots, this paper presents a set of key technical contributions including the incorporation of receding horizon control, visual conditioning, and the time-series diffusion transformer. We hope this work will help motivate a new generation of policy learning techniques that are able to leverage the powerful generative modeling capabilities of diffusion models. Code, data, and training details is publicly available diffusion-policy.cs.columbia.edu
format Preprint
id arxiv_https___arxiv_org_abs_2303_04137
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
Chi, Cheng
Xu, Zhenjia
Feng, Siyuan
Cousineau, Eric
Du, Yilun
Burchfiel, Benjamin
Tedrake, Russ
Song, Shuran
Robotics
This paper introduces Diffusion Policy, a new way of generating robot behavior by representing a robot's visuomotor policy as a conditional denoising diffusion process. We benchmark Diffusion Policy across 12 different tasks from 4 different robot manipulation benchmarks and find that it consistently outperforms existing state-of-the-art robot learning methods with an average improvement of 46.9%. Diffusion Policy learns the gradient of the action-distribution score function and iteratively optimizes with respect to this gradient field during inference via a series of stochastic Langevin dynamics steps. We find that the diffusion formulation yields powerful advantages when used for robot policies, including gracefully handling multimodal action distributions, being suitable for high-dimensional action spaces, and exhibiting impressive training stability. To fully unlock the potential of diffusion models for visuomotor policy learning on physical robots, this paper presents a set of key technical contributions including the incorporation of receding horizon control, visual conditioning, and the time-series diffusion transformer. We hope this work will help motivate a new generation of policy learning techniques that are able to leverage the powerful generative modeling capabilities of diffusion models. Code, data, and training details is publicly available diffusion-policy.cs.columbia.edu
title Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
topic Robotics
url https://arxiv.org/abs/2303.04137