Grasp Diffusion Network: Learning Grasp Generators from Partial Point Clouds with Diffusion Models in SO(3)xR3

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Carvalho, Joao, Le, An T., Jahr, Philipp, Sun, Qiao, Urain, Julen, Koert, Dorothea, Peters, Jan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910740074463232
author Carvalho, Joao
Le, An T.
Jahr, Philipp
Sun, Qiao
Urain, Julen
Koert, Dorothea
Peters, Jan
author_facet Carvalho, Joao
Le, An T.
Jahr, Philipp
Sun, Qiao
Urain, Julen
Koert, Dorothea
Peters, Jan
contents Grasping objects successfully from a single-view camera is crucial in many robot manipulation tasks. An approach to solve this problem is to leverage simulation to create large datasets of pairs of objects and grasp poses, and then learn a conditional generative model that can be prompted quickly during deployment. However, the grasp pose data is highly multimodal since there are several ways to grasp an object. Hence, in this work, we learn a grasp generative model with diffusion models to sample candidate grasp poses given a partial point cloud of an object. A novel aspect of our method is to consider diffusion in the manifold space of rotations and to propose a collision-avoidance cost guidance to improve the grasp success rate during inference. To accelerate grasp sampling we use recent techniques from the diffusion literature to achieve faster inference times. We show in simulation and real-world experiments that our approach can grasp several objects from raw depth images with $90\%$ success rate and benchmark it against several baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2412_08398
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Grasp Diffusion Network: Learning Grasp Generators from Partial Point Clouds with Diffusion Models in SO(3)xR3
Carvalho, Joao
Le, An T.
Jahr, Philipp
Sun, Qiao
Urain, Julen
Koert, Dorothea
Peters, Jan
Robotics
Machine Learning
Grasping objects successfully from a single-view camera is crucial in many robot manipulation tasks. An approach to solve this problem is to leverage simulation to create large datasets of pairs of objects and grasp poses, and then learn a conditional generative model that can be prompted quickly during deployment. However, the grasp pose data is highly multimodal since there are several ways to grasp an object. Hence, in this work, we learn a grasp generative model with diffusion models to sample candidate grasp poses given a partial point cloud of an object. A novel aspect of our method is to consider diffusion in the manifold space of rotations and to propose a collision-avoidance cost guidance to improve the grasp success rate during inference. To accelerate grasp sampling we use recent techniques from the diffusion literature to achieve faster inference times. We show in simulation and real-world experiments that our approach can grasp several objects from raw depth images with $90\%$ success rate and benchmark it against several baselines.
title Grasp Diffusion Network: Learning Grasp Generators from Partial Point Clouds with Diffusion Models in SO(3)xR3
topic Robotics
Machine Learning
url https://arxiv.org/abs/2412.08398