arXivDaily arXiv每日学术速递 周一至周五更新

视觉与机器人

机器人 / 具身智能

机器人、具身智能、机器人学习、操作、导航和具身世界模型。

共收录 61386 信号源:cs.RO, cs.AI, cs.CV, cs.LG

1. 机器人操作 22146 篇

1909.05405 2020-02-20 cs.RO cs.CV eess.IV 89%

SuPer: A Surgical Perception Framework for Endoscopic Tissue Manipulation with Surgical Robotics

Yang Li, Florian Richter, Jingpei Lu, Emily K. Funk, Ryan K. Orosco, Jianke Zhu, Michael C. Yip

专题命中 机器人操作 :manipulation(title,abstract);robotics(title,journal_ref);robotic(abstract);分类 cs.RO、cs.CV

Comments The first two authors made equal contribution on this paper

Journal ref IEEE Robotics and Automation Letters (RA-L), vol. 5, no. 2, pp. 2294-2301, April 2020

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.04884 2026-04-17 cs.RO 89%

Hoi! - A Multimodal Dataset for Force-Grounded, Cross-View Articulated Manipulation

嗨!- 一个用于力导向、跨视角人工 manipulation 的多模态数据集

Tim Engelbracht, René Zurbrügg, Matteo Wohlrapp, Martin Büchner, Abhinav Valada, Marc Pollefeys, Hermann Blum, Zuria Bauer

机构 * ETH Zurich(苏黎世联邦理工学院) Technical University of Munich(慕尼黑技术大学) University of Freiburg(弗赖堡大学) Microsoft(微软) University of Bonn(波恩大学)

专题命中 机器人操作 :manipulation(title,title_cn);robotic(abstract);分类 cs.RO

AI总结 本文提出一个结合视觉、动作和触觉的多模态数据集,包含381个人工物体在38种环境中的3048个序列,用于评估方法在人机视角间的迁移能力及探索交互力等未被研究的模态。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21811 2026-02-26 cs.RO 89%

DexRepNet++: Learning Dexterous Robotic Manipulation with Geometric and Spatial Hand-Object Representations

DexRepNet++: 通过几何和空间手-物体表示学习灵巧的机器人操作

Qingtao Liu, Zhengnan Sun, Yu Cui, Haoming Li, Gaofeng Li, Lin Shao, Jiming Chen, Qi Ye

机构 * IEEE Publication Technology Group(IEEE出版技术组)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);robotics(comments,journal_ref);分类 cs.RO

AI总结 DexRepNet++通过几何和空间手-物体表示提升机器人灵巧操作的泛化能力和效率。

Comments Accepted by IEEE Transactions on Robotics (T-RO), 2026

Journal ref IEEE Transactions on Robotics, vol. 42, pp. 799-818, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2402.07127 2025-10-22 cs.RO cs.AI cs.CV cs.LG 89%

Learning by Watching: A Review of Video-based Learning Approaches for Robot Manipulation

Chrisantus Eze, Christopher Crick

机构 * Computer Science Department, Oklahoma State University, Stillwater, Oklahoma, USA(俄克拉荷马州立大学计算机科学系)

专题命中 机器人操作 :manipulation(title,abstract);robotics(abstract);robot learning(abstract);robotic(abstract)

Comments Published at IEEE Access

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.04445 2025-10-17 cs.RO cs.AI cs.CL cs.CV cs.LG 89%

Moto: Latent Motion Token as the Bridging Language for Learning Robot Manipulation from Videos

Yi Chen, Yuying Ge, Weiliang Tang, Yizhuo Li, Yixiao Ge, Mingyu Ding, Ying Shan, Xihui Liu

机构 * The University of Hong Kong(香港大学) ARC Lab, Tencent PCG(腾讯PCG实验室) The Chinese University of Hong Kong(香港中文大学) University of California, Berkeley(加州大学伯克利分校)

专题命中 机器人操作 :manipulation(title,abstract);robotics(abstract);robot learning(abstract);robotic(abstract)

Comments ICCV 2025. Project page: https://chenyi99.github.io/moto/

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.06048 2025-09-09 cs.RO 89%

Robotic Manipulation Framework Based on Semantic Keypoints for Packing Shoes of Different Sizes, Shapes, and Softness

Yi Dong, Yangjun Liu, Jinjun Duan, Yang Li, Zhendong Dai

机构 * College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics(机械与电子工程学院,南京航空航天大学) State Key Laboratory of Internet of Things for Smart City(智能城市物联网国家重点实验室) Department of Electromechanical Engineering, University of Macau(机电工程系,澳门大学)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);robotics(comments,journal_ref);分类 cs.RO

Comments Yi Dong and Yangjun Liu contributed equally to the work. Accepted by Robotics and Autonomous Systems. https://authors.elsevier.com/c/1lgjX3HdG3supQ

Journal ref Robotics and Autonomous Systems, vol. 194, Dec. 2025, 105174

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.11402 2025-04-03 cs.RO 89%

M2Diffuser: Diffusion-based Trajectory Optimization for Mobile Manipulation in 3D Scenes

Sixu Yan, Zeyu Zhang, Muzhi Han, Zaijin Wang, Qi Xie, Zhitian Li, Zhehan Li, Hangxin Liu, Xinggang Wang, Song-Chun Zhu

专题命中 机器人操作 :manipulation(title,abstract);robotics(abstract);embodied AI(abstract);navigation(abstract)

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.13211 2025-03-03 cs.RO cs.AI cs.CV cs.LG 89%

ManiSkill-HAB: A Benchmark for Low-Level Manipulation in Home Rearrangement Tasks

Arth Shukla, Stone Tao, Hao Su

专题命中 机器人操作 :manipulation(title,abstract);robotics(abstract);embodied AI(abstract);navigation(abstract)

详情

展开后加载摘要…

URL PDF HTML 收藏
2401.12202 2024-11-20 cs.RO cs.AI cs.CV cs.LG 89%

OK-Robot: What Really Matters in Integrating Open-Knowledge Models for Robotics

Peiqi Liu, Yaswanth Orru, Jay Vakil, Chris Paxton, Nur Muhammad Mahi Shafiullah, Lerrel Pinto

专题命中 机器人操作 :robotics(title,abstract);manipulation(abstract);navigation(abstract);robotic(abstract)

Comments Github repo: https://github.com/ok-robot/ok-robot

详情

展开后加载摘要…

URL PDF HTML 收藏
2405.08576 2024-05-15 cs.RO cs.AI cs.CV cs.LG 89%

Hearing Touch: Audio-Visual Pretraining for Contact-Rich Manipulation

Jared Mejia, Victoria Dean, Tess Hellebrekers, Abhinav Gupta

专题命中 机器人操作 :manipulation(title,abstract);robotics(abstract);robot learning(abstract);robotic(abstract)

Comments Accepted to ICRA 2024

详情

展开后加载摘要…

URL PDF HTML 收藏
2401.10702 2024-01-22 cs.RO 89%

G.O.G: A Versatile Gripper-On-Gripper Design for Bimanual Cloth Manipulation with a Single Robotic Arm

Dongmyoung Lee, Wei Chen, Xiaoshuai Chen, Nicolas Rojas

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);robotics(comments,journal_ref);分类 cs.RO

Comments Accepted for IEEE Robotics and Automation Letters in January 2024. Dongmyoung Lee and Wei Chen contributed equally to this research

Journal ref IEEE Robotics and Automation Letters 2024

详情

展开后加载摘要…

URL PDF HTML 收藏
2104.11213 2021-04-23 cs.CV cs.AI cs.LG cs.RO 89%

ManipulaTHOR: A Framework for Visual Object Manipulation

Kiana Ehsani, Winson Han, Alvaro Herrasti, Eli VanderBilt, Luca Weihs, Eric Kolve, Aniruddha Kembhavi, Roozbeh Mottaghi

专题命中 机器人操作 :manipulation(title,abstract);robotics(abstract);embodied AI(abstract);navigation(abstract)

Comments CVPR 2021 -- (Oral presentation)

详情

展开后加载摘要…

URL PDF HTML 收藏
2010.06544 2021-03-02 cs.RO 89%

Real-Time Deep Learning Approach to Visual Servo Control and Grasp Detection for Autonomous Robotic Manipulation

Eduardo Godinho Ribeiro, Raul de Queiroz Mendes, Valdir Grassi

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);robotics(comments,journal_ref);分类 cs.RO

Comments 27 pages, 24 figures. Preprint published in the Elsevier's Robotics and Autonomous Systems journal on February 24, 2021

Journal ref Journal: Robotics and Autonomous Systems, publisher: Elsevier, volume number: 139, year: 2021, page number: 103757

详情

展开后加载摘要…

URL PDF HTML 收藏
2303.00905 2023-10-27 cs.RO cs.AI cs.CV 89%

Open-World Object Manipulation using Pre-trained Vision-Language Models

Austin Stone, Ted Xiao, Yao Lu, Keerthana Gopalakrishnan, Kuang-Huei Lee, Quan Vuong, Paul Wohlhart, Sean Kirmani, Brianna Zitkovich, Fei Xia, Chelsea Finn, Karol Hausman

专题命中 机器人操作 :manipulation(title,abstract);robot learning(abstract,comments);navigation(abstract);robot policy(abstract)

Comments Accepted at the 7th Conference on Robot Learning (CoRL 2023)

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.22296 2026-08-25 cs.RO cs.CV 新提交 88%

TONAV: Task-Oriented Navigation and Action-Velocity Chunk Learning for Articulated Object Quadrupedal Mobile Manipulation

TONAV:面向关节物体四足移动操作的任务导向导航与动作-速度块学习

Haoran Lin, Mingyu Yang, Pengfei Qi, Kehan Chen, Qiang Diao, Liangji Zeng, Wenrui Chen, Yaonan Wang, Kailun Yang

机构 * School of Artificial Intelligence and Robotics, Hunan University(湖南大学人工智能与机器人学院) National Engineering Research Center of Robot Visual Perception and Control Technology, Hunan University(湖南大学机器人视觉感知与控制技术国家工程研究中心) College of Semiconductors (College of Integrated Circuits), Hunan University(湖南大学半导体学院(集成电路学院))

专题命中 机器人操作 :manipulation(title,abstract);navigation(title,abstract);分类 cs.RO、cs.CV

AI总结 该研究针对四足移动操作中导航与操作的耦合问题,提出TONAV框架,结合任务导向导航与动作-速度块学习,通过遥操作、视觉语言推理及速度监督实现更优的操作性能。

Comments The project page is at https://haochen611.github.io/TONAV

详情

展开后加载摘要…

URL PDF HTML 收藏
2406.01586 2026-08-19 cs.RO cs.AI 版本更新 88%

ManiCM: Real-time 3D Diffusion Policy via Consistency Model for Robotic Manipulation

ManiCM:用于机器人操作的基于一致性模型的实时3D扩散策略

Zifeng Gao, Guanxing Lu, Tianxing Chen, Wenxun Dai, Ziwei Wang, Chao Shang, Wenbo Ding, Yansong Tang

机构 * Tsinghua Shenzhen International Graduate School, Tsinghua University(清华大学深圳国际研究生院,清华大学) Shanghai AI Laboratory(上海人工智能实验室) Nanyang Technological University(南洋理工大学)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 该研究提出ManiCM模型,通过一致性约束实现一步推理,在31项机器人操作任务上,推理速度较最优方法提升10倍且保持竞争力。

Comments https://manicm-fast.github.io/

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.14986 2026-08-18 cs.RO cs.AI 新提交 88%

GaussMemory: Task-Driven 3D Gaussian Scene Memory for Long-Horizon Robotic Manipulation

GaussMemory:面向长时程机器人操作的任务驱动三维高斯场景记忆

Zhiqiang Hu, Shouren Huang, Masatoshi Ishikawa

机构 * Research Institute for Science & Technology, Tokyo University of Science(东京理科大学科学技术研究所)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 该研究针对现有三维机器人记忆系统被动存储的缺陷,提出任务驱动的主动场景记忆框架GaussMemory,在LIBERO、VLABench基准上超越了MemoryVLA、π₀-FAST等方法。

Comments 8 pages, 10 figures. Accepted to the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.01229 2026-07-31 cs.RO cs.AI 版本更新 88%

RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Design

RMBench: 具有政策设计洞察的内存依赖机器人操作基准

Tianxing Chen, Yuran Wang, Mingleyang Li, Yan Qin, Hao Shi, Zixuan Li, Yifan Hu, Yingsheng Zhang, Kaixuan Wang, Yue Chen, Hongcheng Wang, Junjie Wang, Tianhang Yang, Renjing Xu, Ruihai Wu, Yao Mu, Yaodong Yang, Hao Dong, Ping Luo

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 RMBench通过9个涵盖多层级内存复杂度的任务系统评估政策内存能力,提出Mem-0模块化政策支持可控消融研究,揭示现有政策的内存限制及架构设计对内存性能的影响。

Comments website: https://rmbench.github.io/

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.13653 2026-07-28 cs.CV cs.RO 版本更新 88%

Exploratory, Communicative, and Deployable: Vision-Driven Embodied Agents for Open-World Mobile Manipulation

探索性、交流性和可部署性:用于开放世界移动操作的视觉驱动具身智能体

Boyu Mi, Mengchen Ma, Yifei Yao, Xing Gao, Junting Chen, Yangzi Li, Zihou Zhu, Guohao Li, Zhenfei Yin, Tai Wang, Yao Mu, Jiangmiao Pang, Hanqing Wang

专题命中 机器人操作 :embodied agent(title,abstract);manipulation(title,abstract);分类 cs.RO、cs.CV

AI总结 研究针对具身智能体现实部署难题,提出REAL框架,通过建立一致环境API、设计任务组合优化性能,引入REAL - Bench评估。实验显示其智能体在交互式任务中表现出色,在物理机器人上也实现高成功率,弥合现实差距。

Comments Accepted to ECCV 2026. 57 pages. Code available at https://github.com/InternRobotics/REAL

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.21378 2026-07-22 cs.RO cs.AI 版本更新 88%

RoboInspector: Unveiling the Unreliability of Policy Code for LLM-enabled Robotic Manipulation

RoboInspector:揭示用于基于大语言模型的机器人操作的策略代码的不可靠性

Chenduo Ying, Linkang Du, Peng Cheng, Yuanchao Shu

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 研究基于大语言模型的机器人操作中策略代码不可靠问题,设计RoboInspector从任务复杂性和指令粒度揭示其不可靠性,通过实验识别四种不可靠行为及原因,还引入优化方法,可提高策略代码生成可靠性达35%。

Comments Accepted to ACM Transactions on Intelligent Systems and Technology

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.17861 2026-07-21 cs.RO cs.AI 新提交 88%

ConceptTree: Bringing Semantic Transparency to Black-Box Decision Making for Robotic Manipulation

ConceptTree:为机器人操作的黑箱决策带来语义透明度

Yongyan Wen, Feifan Liu, Jinyi Chen, Bo An, Peng Liu, Siyuan Li

机构 * Harbin Institute of Technology(哈尔滨工业大学) Nanyang Technological University(南洋理工大学)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 研究长期机器人操作中可解释决策问题,提出ConceptTree框架,将高级技能选择重构为基于视觉观察的概念推理。该方法学习归一化概念空间并训练决策树,实验表明其优于现有基线,支持细粒度干预以纠正决策错误。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.17323 2026-07-21 cs.RO cs.CV cs.HC cs.SY eess.SY 新提交 88%

From Perception to Assistance: Open-Vocabulary Shared Autonomy for Robotic Manipulation

从感知到协助:用于机器人操作的开放词汇共享自主性

Murilo Vinicius da Silva, Ricardo V. Godoy, Juliano Negri, Gustavo J. G. Lahr, Ranulfo Bezerra, Marcelo Becker

机构 * University of São Paulo(圣保罗大学) Instituto Israelita de Ensino e Pesquisa Albert Einstein(以色列爱因斯坦教育与研究机构) Graduate School of Information Sciences, Tohoku University(东北大学信息科学研究生院)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.CV

AI总结 研究针对工业环境中机器人操作难题,提出共享自主性框架,利用RGB-D摄像头、视觉语言模型等,经GPU加速控制器等实现操作,在四足移动操纵器上验证,在多项任务中表现良好,展示了该框架的有效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.14005 2026-07-16 cs.CV cs.RO 新提交 88%

M$^\text{4}$World: A Multi-view Multimodal Driving World Model for Interactive Object Manipulation and Minute-long Streaming

M$^\text{4}$World:用于交互式对象操纵和分钟级流的多视图多模态驾驶世界模型

Ke Cheng, Hanqiao Ye, Lei Shi, Yahui Liu, Yunhan Shen, Jingtao Dong, Zhenke Wang, Wenxuan Ao, Weixiang Xu, Kaining Huang, Shuhan Shen

专题命中 机器人操作 :manipulation(title,abstract);world model(title,abstract);分类 cs.RO、cs.CV

AI总结 针对现有驾驶世界生成方法局限,提出M$^\text{4}$World模型,通过灵活接口与多阶段训练实现对象操纵及长时流稳定,引入后训练与生成模型,并用新管道评估,实验证明其在驾驶模拟中有高质量、可控性与稳定性。

Comments 24 pages, 13 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.01067 2026-07-02 cs.RO cs.CV 新提交 88%

Human-Centric Transferable Tactile Pre-Training for Dexterous Robotic Manipulation

面向灵巧机器人操作的人为中心的可迁移触觉预训练

Chi Zhang, Penglin Cai, Ziheng Xi, Haoqi Yuan, Hao Luo, Wanpeng Zhang, Sipeng Zheng, Chaoyi Xu, Zongqing Lu

机构 * Peking University(北京大学) BeingBeyond Tsinghua University(清华大学)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.CV

AI总结 提出H-Tac大规模触觉-动作数据集和可迁移触觉预训练(TTP)系统,通过统一触觉与动作空间弥合人机差距,利用触觉专家预测未来触觉以建模接触动力学,显著提升灵巧操作的泛化与精细控制能力。

Comments The first two authors contribute equally. Orders are decided by flipping a coin

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.31494 2026-07-01 cs.RO cs.AI 新提交 88%

Robustness of Robotic Manipulation: Foundations and Frontiers

机器人操作的鲁棒性:基础与前沿

Yifei Dong, Zhanyi Sun, Lujie Yang, Manuel Baum, Kei Ikemura, Shuran Song, Florian T. Pokorny, Xianyi Cheng

机构 * Duke University(杜克大学) KTH Royal Institute of Technology(瑞典皇家理工学院) Stanford University(斯坦福大学) Massachusetts Institute of Technology(麻省理工学院) Organifarms

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 本文系统研究机器人操作鲁棒性,提出正式定义、概率与控制理论框架,并综述感知、规划、控制、策略学习和硬件中的鲁棒性机制及评估方法,展望实现人类级鲁棒性的未来方向。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.27268 2026-06-26 cs.RO cs.AI 新提交 88%

E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulation

E-TTS:一种新的机器人操作具身测试时缩放框架

Wen Ye, Peiyan Li, Tingyu Yuan, Yuan Xu, Xiangnan Wu, Chaoyang Zhao, Jing Liu, Nianfeng Liu, Yan Huang, Liang Wang

机构 * New Laboratory of Pattern Recognition (NLPR), Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所模式识别国家重点实验室) School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院) Foundation Model Research Center, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所基础模型研究中心) FiveAges(五时代)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 提出E-TTS框架,通过历史感知迭代精炼和视觉语言验证器统一推理与动作缩放,解决具身任务中推理缩放和历史信息利用不足的问题,在仿真和真实场景中分别提升33.14%和26.62%的性能。

Comments Accepted to ECCV 2026. 44 pages, 11 figures. Project page: https://27yw.github.io/E-TTS-Web/

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.00324 2026-06-24 cs.RO cs.CV cs.HC 版本更新 88%

MILE: A Mechanically Isomorphic Hand Exoskeleton and Visuotactile Robotic Hand for Data Collection in Dexterous Manipulation

MILE:一种用于灵巧操作的指尖视觉触觉传感的机械同构外骨骼数据采集系统

Jinda Du, Jieji Ren, Qiaojun Yu, Ningbin Zhang, Yu Deng, Xingyu Wei, Yufei Liu, Guoying Gu, Xiangyang Zhu

机构 * State Key Laboratory of Mechanical System and Vibration, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China(机械系统与振动国家重点实验室,上海交通大学机械工程学院,上海200240,中国) Shanghai Key Laboratory of Intelligent Robotics, Shanghai Jiao Tong University, Shanghai 200240, China(智能机器人上海重点实验室,上海交通大学,上海200240,中国) Shanghai Artificial Intelligence Laboratory, Shanghai, China(上海人工智能实验室,上海,中国) Humanoid Robot (Shanghai) Co., Ltd., Shanghai, China(上海人形机器人有限公司,上海,中国)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.CV

AI总结 针对现有灵巧操作数据采集系统运动重定向不准确、效率低、缺乏高分辨率指尖触觉传感的问题,提出MILE系统,通过从人手到外骨骼再到机械手的机械同构设计,实现无重定向精确控制,并集成指尖视觉触觉模块,采集多模态数据集,显著提升遥操作成功率。

Comments 18 pages including supplementary material. Main manuscript and supplementary material included in this version

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.05226 2026-06-24 cs.RO cs.AI 版本更新 88%

Reward-Centered ReST-MCTS: A Robust Decision-Making Framework for Robotic Manipulation in High Uncertainty Environments

以奖励为中心的ReST-MCTS:高不确定性环境下机器人操作的鲁棒决策框架

Xibai Wang

机构 * Xibai Wang(王西拜)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.AI

AI总结 提出Reward-Centered ReST-MCTS框架,通过分解中间反馈为中心信号并引导搜索,解决高不确定性环境下MCTS因稀疏奖励和噪声导致的决策脆弱性问题,实验验证其在同骨干VLA上的鲁棒性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.10818 2026-06-10 cs.RO cs.CV 新提交 88%

IMPACT: Learning Internal-Model Predictive Control for Forceful Robotic Manipulation

IMPACT:面向强力机器人操控的内部模型预测控制学习

Jiawei Gao, Chaoqi Liu, Peilin Wu, Haonan Chen, Yilun Du

机构 * Harvard University(哈佛大学) Stanford University(斯坦福大学)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.CV

AI总结 提出IMPACT框架,将强力操控任务解耦为任务规划和基于内部模型的预测控制,通过仿真和实验证明其在成功率、泛化性、安全性和能效上的优势。

Comments Project website: https://gao-jiawei.com/IMPACT/

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.09989 2026-06-05 cs.RO cs.CV 88%

StereoPolicy: Improving Robotic Manipulation Policies via Stereo Perception

StereoPolicy:通过立体视觉改进机器人操作策略

Evans Han, Yunfan Jiang, Yingke Wang, Haoyue Xiao, Huang Huang, Jianwen Xie, Jiajun Wu, Li Fei-Fei, Ruohan Zhang

机构 * Stanford University(斯坦福大学) Northwestern University(西北大学) Lambda, Inc(Lambda公司)

专题命中 机器人操作 :manipulation(title,abstract);robotic(title,abstract);分类 cs.RO、cs.CV

AI总结 该研究提出StereoPolicy,一种利用立体视觉提升机器人操作策略的框架,通过同步立体图像对增强几何推理,无需构建显式3D表示,在多个仿真和真实机器人任务中优于RGB、RGB-D、点云等基线方法。

详情

展开后加载摘要…

URL PDF HTML 收藏