Robo-GS: A Physics Consistent Spatial-Temporal Model for Robotic Arm with Hybrid Representation
Real2Sim2Real plays a critical role in robotic arm control and reinforcement learning, yet bridging this gap remains a significant challenge due to the complex physical properties of robots and the objects they manipulate. Existing methods lack a comprehensive solution to accurately reconstruct real...
Gespeichert in:
Hauptverfasser: | , , , , , , , , , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | |
container_volume | |
creator | Lou, Haozhe Liu, Yurong Pan, Yike Geng, Yiran Chen, Jianteng Ma, Wenlong Li, Chenglong Wang, Lin Feng, Hengzhen Shi, Lu Luo, Liyi Shi, Yongliang |
description | Real2Sim2Real plays a critical role in robotic arm control and reinforcement
learning, yet bridging this gap remains a significant challenge due to the
complex physical properties of robots and the objects they manipulate. Existing
methods lack a comprehensive solution to accurately reconstruct real-world
objects with spatial representations and their associated physics attributes.
We propose a Real2Sim pipeline with a hybrid representation model that
integrates mesh geometry, 3D Gaussian kernels, and physics attributes to
enhance the digital asset representation of robotic arms.
This hybrid representation is implemented through a Gaussian-Mesh-Pixel
binding technique, which establishes an isomorphic mapping between mesh
vertices and Gaussian models. This enables a fully differentiable rendering
pipeline that can be optimized through numerical solvers, achieves
high-fidelity rendering via Gaussian Splatting, and facilitates physically
plausible simulation of the robotic arm's interaction with its environment
using mesh-based methods.
The code,full presentation and datasets will be made publicly available at
our website https://robostudioapp.com |
doi_str_mv | 10.48550/arxiv.2408.14873 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2408_14873</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2408_14873</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2408_148733</originalsourceid><addsrcrecordid>eNqFzrEOgjAUheEuDkZ9ACfvC4AgEIkbISqLiQH2pkAJNym0aRuVtxeIu9NZzp98hOx9zw3jKPKOTH_w5Z5CL3b9MD4Ha0JzWUnnXlwggWc3GqwNpHIwaCwfLBSKWWTCKXmvpGYCHrLhAlqpYQ4t1pDoHt5oO8jGSmMDOVeamymeSjlsyaplwvDdbzfkcLuWaeYsFKo09kyPdCbRhRT8f3wBpzJBnQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Robo-GS: A Physics Consistent Spatial-Temporal Model for Robotic Arm with Hybrid Representation</title><source>arXiv.org</source><creator>Lou, Haozhe ; Liu, Yurong ; Pan, Yike ; Geng, Yiran ; Chen, Jianteng ; Ma, Wenlong ; Li, Chenglong ; Wang, Lin ; Feng, Hengzhen ; Shi, Lu ; Luo, Liyi ; Shi, Yongliang</creator><creatorcontrib>Lou, Haozhe ; Liu, Yurong ; Pan, Yike ; Geng, Yiran ; Chen, Jianteng ; Ma, Wenlong ; Li, Chenglong ; Wang, Lin ; Feng, Hengzhen ; Shi, Lu ; Luo, Liyi ; Shi, Yongliang</creatorcontrib><description>Real2Sim2Real plays a critical role in robotic arm control and reinforcement
learning, yet bridging this gap remains a significant challenge due to the
complex physical properties of robots and the objects they manipulate. Existing
methods lack a comprehensive solution to accurately reconstruct real-world
objects with spatial representations and their associated physics attributes.
We propose a Real2Sim pipeline with a hybrid representation model that
integrates mesh geometry, 3D Gaussian kernels, and physics attributes to
enhance the digital asset representation of robotic arms.
This hybrid representation is implemented through a Gaussian-Mesh-Pixel
binding technique, which establishes an isomorphic mapping between mesh
vertices and Gaussian models. This enables a fully differentiable rendering
pipeline that can be optimized through numerical solvers, achieves
high-fidelity rendering via Gaussian Splatting, and facilitates physically
plausible simulation of the robotic arm's interaction with its environment
using mesh-based methods.
The code,full presentation and datasets will be made publicly available at
our website https://robostudioapp.com</description><identifier>DOI: 10.48550/arxiv.2408.14873</identifier><language>eng</language><subject>Computer Science - Numerical Analysis ; Computer Science - Robotics ; Mathematics - Numerical Analysis ; Mathematics - Optimization and Control</subject><creationdate>2024-08</creationdate><rights>http://creativecommons.org/licenses/by-nc-sa/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2408.14873$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2408.14873$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Lou, Haozhe</creatorcontrib><creatorcontrib>Liu, Yurong</creatorcontrib><creatorcontrib>Pan, Yike</creatorcontrib><creatorcontrib>Geng, Yiran</creatorcontrib><creatorcontrib>Chen, Jianteng</creatorcontrib><creatorcontrib>Ma, Wenlong</creatorcontrib><creatorcontrib>Li, Chenglong</creatorcontrib><creatorcontrib>Wang, Lin</creatorcontrib><creatorcontrib>Feng, Hengzhen</creatorcontrib><creatorcontrib>Shi, Lu</creatorcontrib><creatorcontrib>Luo, Liyi</creatorcontrib><creatorcontrib>Shi, Yongliang</creatorcontrib><title>Robo-GS: A Physics Consistent Spatial-Temporal Model for Robotic Arm with Hybrid Representation</title><description>Real2Sim2Real plays a critical role in robotic arm control and reinforcement
learning, yet bridging this gap remains a significant challenge due to the
complex physical properties of robots and the objects they manipulate. Existing
methods lack a comprehensive solution to accurately reconstruct real-world
objects with spatial representations and their associated physics attributes.
We propose a Real2Sim pipeline with a hybrid representation model that
integrates mesh geometry, 3D Gaussian kernels, and physics attributes to
enhance the digital asset representation of robotic arms.
This hybrid representation is implemented through a Gaussian-Mesh-Pixel
binding technique, which establishes an isomorphic mapping between mesh
vertices and Gaussian models. This enables a fully differentiable rendering
pipeline that can be optimized through numerical solvers, achieves
high-fidelity rendering via Gaussian Splatting, and facilitates physically
plausible simulation of the robotic arm's interaction with its environment
using mesh-based methods.
The code,full presentation and datasets will be made publicly available at
our website https://robostudioapp.com</description><subject>Computer Science - Numerical Analysis</subject><subject>Computer Science - Robotics</subject><subject>Mathematics - Numerical Analysis</subject><subject>Mathematics - Optimization and Control</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNqFzrEOgjAUheEuDkZ9ACfvC4AgEIkbISqLiQH2pkAJNym0aRuVtxeIu9NZzp98hOx9zw3jKPKOTH_w5Z5CL3b9MD4Ha0JzWUnnXlwggWc3GqwNpHIwaCwfLBSKWWTCKXmvpGYCHrLhAlqpYQ4t1pDoHt5oO8jGSmMDOVeamymeSjlsyaplwvDdbzfkcLuWaeYsFKo09kyPdCbRhRT8f3wBpzJBnQ</recordid><startdate>20240827</startdate><enddate>20240827</enddate><creator>Lou, Haozhe</creator><creator>Liu, Yurong</creator><creator>Pan, Yike</creator><creator>Geng, Yiran</creator><creator>Chen, Jianteng</creator><creator>Ma, Wenlong</creator><creator>Li, Chenglong</creator><creator>Wang, Lin</creator><creator>Feng, Hengzhen</creator><creator>Shi, Lu</creator><creator>Luo, Liyi</creator><creator>Shi, Yongliang</creator><scope>AKY</scope><scope>AKZ</scope><scope>GOX</scope></search><sort><creationdate>20240827</creationdate><title>Robo-GS: A Physics Consistent Spatial-Temporal Model for Robotic Arm with Hybrid Representation</title><author>Lou, Haozhe ; Liu, Yurong ; Pan, Yike ; Geng, Yiran ; Chen, Jianteng ; Ma, Wenlong ; Li, Chenglong ; Wang, Lin ; Feng, Hengzhen ; Shi, Lu ; Luo, Liyi ; Shi, Yongliang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2408_148733</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Numerical Analysis</topic><topic>Computer Science - Robotics</topic><topic>Mathematics - Numerical Analysis</topic><topic>Mathematics - Optimization and Control</topic><toplevel>online_resources</toplevel><creatorcontrib>Lou, Haozhe</creatorcontrib><creatorcontrib>Liu, Yurong</creatorcontrib><creatorcontrib>Pan, Yike</creatorcontrib><creatorcontrib>Geng, Yiran</creatorcontrib><creatorcontrib>Chen, Jianteng</creatorcontrib><creatorcontrib>Ma, Wenlong</creatorcontrib><creatorcontrib>Li, Chenglong</creatorcontrib><creatorcontrib>Wang, Lin</creatorcontrib><creatorcontrib>Feng, Hengzhen</creatorcontrib><creatorcontrib>Shi, Lu</creatorcontrib><creatorcontrib>Luo, Liyi</creatorcontrib><creatorcontrib>Shi, Yongliang</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv Mathematics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Lou, Haozhe</au><au>Liu, Yurong</au><au>Pan, Yike</au><au>Geng, Yiran</au><au>Chen, Jianteng</au><au>Ma, Wenlong</au><au>Li, Chenglong</au><au>Wang, Lin</au><au>Feng, Hengzhen</au><au>Shi, Lu</au><au>Luo, Liyi</au><au>Shi, Yongliang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Robo-GS: A Physics Consistent Spatial-Temporal Model for Robotic Arm with Hybrid Representation</atitle><date>2024-08-27</date><risdate>2024</risdate><abstract>Real2Sim2Real plays a critical role in robotic arm control and reinforcement
learning, yet bridging this gap remains a significant challenge due to the
complex physical properties of robots and the objects they manipulate. Existing
methods lack a comprehensive solution to accurately reconstruct real-world
objects with spatial representations and their associated physics attributes.
We propose a Real2Sim pipeline with a hybrid representation model that
integrates mesh geometry, 3D Gaussian kernels, and physics attributes to
enhance the digital asset representation of robotic arms.
This hybrid representation is implemented through a Gaussian-Mesh-Pixel
binding technique, which establishes an isomorphic mapping between mesh
vertices and Gaussian models. This enables a fully differentiable rendering
pipeline that can be optimized through numerical solvers, achieves
high-fidelity rendering via Gaussian Splatting, and facilitates physically
plausible simulation of the robotic arm's interaction with its environment
using mesh-based methods.
The code,full presentation and datasets will be made publicly available at
our website https://robostudioapp.com</abstract><doi>10.48550/arxiv.2408.14873</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.2408.14873 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_2408_14873 |
source | arXiv.org |
subjects | Computer Science - Numerical Analysis Computer Science - Robotics Mathematics - Numerical Analysis Mathematics - Optimization and Control |
title | Robo-GS: A Physics Consistent Spatial-Temporal Model for Robotic Arm with Hybrid Representation |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-28T03%3A01%3A31IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Robo-GS:%20A%20Physics%20Consistent%20Spatial-Temporal%20Model%20for%20Robotic%20Arm%20with%20Hybrid%20Representation&rft.au=Lou,%20Haozhe&rft.date=2024-08-27&rft_id=info:doi/10.48550/arxiv.2408.14873&rft_dat=%3Carxiv_GOX%3E2408_14873%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |