Target-Oriented Object Grasping via Multimodal Human Guidance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xie, Pengwei, Chen, Siang, Hu, Dingchang, Dai, Yixiang, Yang, Kaiqin, Wang, Guijin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914917989220352
author Xie, Pengwei
Chen, Siang
Hu, Dingchang
Dai, Yixiang
Yang, Kaiqin
Wang, Guijin
author_facet Xie, Pengwei
Chen, Siang
Hu, Dingchang
Dai, Yixiang
Yang, Kaiqin
Wang, Guijin
contents In the context of human-robot interaction and collaboration scenarios, robotic grasping still encounters numerous challenges. Traditional grasp detection methods generally analyze the entire scene to predict grasps, leading to redundancy and inefficiency. In this work, we reconsider 6-DoF grasp detection from a target-referenced perspective and propose a Target-Oriented Grasp Network (TOGNet). TOGNet specifically targets local, object-agnostic region patches to predict grasps more efficiently. It integrates seamlessly with multimodal human guidance, including language instructions, pointing gestures, and interactive clicks. Thus our system comprises two primary functional modules: a guidance module that identifies the target object in 3D space and TOGNet, which detects region-focal 6-DoF grasps around the target, facilitating subsequent motion planning. Through 50 target-grasping simulation experiments in cluttered scenes, our system achieves a success rate improvement of about 13.7%. In real-world experiments, we demonstrate that our method excels in various target-oriented grasping scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11138
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Target-Oriented Object Grasping via Multimodal Human Guidance
Xie, Pengwei
Chen, Siang
Hu, Dingchang
Dai, Yixiang
Yang, Kaiqin
Wang, Guijin
Robotics
Computer Vision and Pattern Recognition
In the context of human-robot interaction and collaboration scenarios, robotic grasping still encounters numerous challenges. Traditional grasp detection methods generally analyze the entire scene to predict grasps, leading to redundancy and inefficiency. In this work, we reconsider 6-DoF grasp detection from a target-referenced perspective and propose a Target-Oriented Grasp Network (TOGNet). TOGNet specifically targets local, object-agnostic region patches to predict grasps more efficiently. It integrates seamlessly with multimodal human guidance, including language instructions, pointing gestures, and interactive clicks. Thus our system comprises two primary functional modules: a guidance module that identifies the target object in 3D space and TOGNet, which detects region-focal 6-DoF grasps around the target, facilitating subsequent motion planning. Through 50 target-grasping simulation experiments in cluttered scenes, our system achieves a success rate improvement of about 13.7%. In real-world experiments, we demonstrate that our method excels in various target-oriented grasping scenarios.
title Target-Oriented Object Grasping via Multimodal Human Guidance
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.11138