机器人 / 具身智能 - arXivDaily 专题

2509.23021 2026-03-17 cs.RO cs.CV 79%

UniPrototype: Humn-Robot Skill Learning with Uniform Prototypes

UniPrototype：基于统一原型的人机技能学习

Xiao Hu, Qi Yin, Yangming Shi, Yang Ye

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV

AI总结本文提出UniPrototype框架，通过共享运动原语实现人机知识迁移，解决机器人学习数据稀缺问题，提升学习效率和任务性能。

Comments This submission was uploaded in error and has been withdrawn. A substantial revision will need to be completed

2512.06013 2026-02-02 cs.CV cs.RO 79%

VAT: Vision Action Transformer by Unlocking Full Representation of ViT

通过解锁ViT的完整表示构建视觉动作变换器

Wenhao Li, Chengwei Ma, Weixin Mao

机构 * Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV

AI总结 VAT通过解锁ViT的完整表示，提出了一种新的视觉动作变换器架构，实现了在模拟操作任务中98.15%的成功率，优于现有方法。

2510.27114 2025-11-03 cs.RO cs.LG 79%

Learning Generalizable Visuomotor Policy through Dynamics-Alignment

Dohyeok Lee, Jung Min Lee, Munkyung Kim, Seokhun Ju, Jin Woo Koo, Kyungjae Lee, Dohyeong Kim, TaeHyun Cho, Jungwoo Lee

机构 * Department of Electrical and Computer Engineering, Seoul National University（电子与计算机工程系，首尔国立大学）； Department of Statistics, Korea University（统计系，韩国大学）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.LG

Comments 9 pages, 6 figures

2503.02048 2025-03-05 cs.RO cs.AI 79%

FRMD: Fast Robot Motion Diffusion with Consistency-Distilled Movement Primitives for Smooth Action Generation

Xirui Shi, Jun Jin

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI

Comments arXiv admin note: text overlap with arXiv:2406.01586 by other authors

2409.03996 2024-09-09 cs.LG cs.RO 79%

Goal-Reaching Policy Learning from Non-Expert Observations via Effective Subgoal Guidance

RenMing Huang, Shaochong Liu, Yunqiang Pei, Peng Wang, Guoqing Wang, Yang Yang, Hengtao Shen

专题命中机器人学习：manipulation(abstract)；navigation(abstract)；robotic(abstract)；分类 cs.RO、cs.LG

Comments Accepted to CoRL 2024

2103.07938 2021-03-16 cs.RO cs.AI 79%

Learning needle insertion from sample task executions

Amir Ghalamzan-E

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI

Comments Submitted to IEEE/RSJ IROS 2021!

2003.08515 2020-03-20 cs.CV cs.RO 79%

SAPIEN: A SimulAted Part-based Interactive ENvironment

Fanbo Xiang, Yuzhe Qin, Kaichun Mo, Yikuan Xia, Hao Zhu, Fangchen Liu, Minghua Liu, Hanxiao Jiang, Yifu Yuan, He Wang, Li Yi, Angel X. Chang, Leonidas J. Guibas, Hao Su

专题命中机器人学习：robotics(abstract)；robot learning(abstract)；robotic(abstract)；分类 cs.RO、cs.CV

1507.01273 2015-09-24 cs.LG cs.RO 79%

Learning Deep Neural Network Policies with Continuous Memory States

Marvin Zhang, Zoe McCarthy, Chelsea Finn, Sergey Levine, Pieter Abbeel

专题命中机器人学习：manipulation(abstract)；navigation(abstract)；robotic(abstract)；分类 cs.RO、cs.LG

2504.21841 2026-02-26 cs.RO cs.FL 78%

Neuro-Symbolic Generation of Explanations for Robot Policies with Weighted Signal Temporal Logic

神经符号生成机器人策略的解释性说明以加权信号时序逻辑

Mikihisa Yuasa, Ramavarapu S. Sreenivas, Huy T. Tran

机构 * The Grainger College of Engineering, University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校格拉inger工程学院）

专题命中机器人学习：robotics(abstract,journal_ref)；robot policy(abstract)；robotic(abstract)；分类 cs.RO

AI总结本文提出神经符号生成方法，通过加权信号时序逻辑生成简洁、一致且严格的解释，提升机器人策略的可解释性和安全性。

Journal ref IEEE Robotics and Automation Letters, vol. 11, pp. 3963-3970, 2026

2405.18196 2024-05-29 cs.RO cs.AI cs.CV cs.LG 78%

Render and Diffuse: Aligning Image and Action Spaces for Diffusion-based Behaviour Cloning

Vitalis Vosylius, Younggyo Seo, Jafar Uruç, Stephen James

专题命中机器人学习：robot learning(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV；robotics(comments)

Comments Robotics: Science and Systems (RSS) 2024. Videos are available on our project webpage at https://vv19.github.io/render-and-diffuse/

2109.09690 2021-09-22 cs.RO cs.AI cs.CV cs.LG 78%

Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes

Jongseok Lee, Jianxiang Feng, Matthias Humt, Marcus G. Müller, Rudolph Triebel

专题命中机器人学习：robot learning(abstract,comments)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments 12 pages, 6 figures and 1 table. Accepted at the 5th Conference on Robot Learning (CORL 2021), London, UK

1903.01973 2019-12-23 cs.RO 78%

Learning Latent Plans from Play

Corey Lynch, Mohi Khansari, Ted Xiao, Vikash Kumar, Jonathan Tompson, Sergey Levine, Pierre Sermanet

专题命中机器人学习：robotics(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

Comments Published at CoRL 2019 (3rd Conference on Robot Learning, Osaka, Japan)

2412.21088 2024-12-31 cs.MA 78%

Advances in Multi-agent Reinforcement Learning: Persistent Autonomy and Robot Learning Lab Report 2024

Reza Azadeh

专题命中机器人学习：robot learning(title,abstract)

Comments 7 pages, 8 figures, technical report

2410.21795 2024-11-05 cs.AI cs.LG cs.RO 78%

Robot Policy Learning with Temporal Optimal Transport Reward

Yuwei Fu, Haichao Zhang, Di Wu, Wei Xu, Benoit Boulet

专题命中机器人学习：robot policy(title)；分类 cs.RO、cs.AI、cs.LG

Comments NeurIPS 2024

2606.15909 2026-06-16 cs.RO 新提交 77%

GeoTLM: Geometry-aware Tactile-Language Models for Contact Motion Orientation Reasoning of Dynamic Objects

GeoTLM: 面向动态物体接触运动方向推理的几何感知触觉语言模型

Qiutian Li, Zinan Liu, Lin Wang

机构 * School of EEE, Nanyang Technological University (NTU)（南洋理工大学电气与电子工程学院）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

AI总结提出GeoTLM，通过可微几何表示（DGR）提取触觉剪切场中的几何先验，提升动态物体旋转和滑动方向推理能力，在旋转和滑动任务上分别提升14.6%和16.2%的准确率。

Comments 7 pages, 3 figures, 4 tables

2605.29766 2026-05-29 cs.RO 77%

MARS Policy: Multimodality Only When It Matters

MARS策略：仅在必要时使用多模态

Jindou Jia, Tuo An, Yuxuan Hu, Gen Li, Jingliang Li, Bohan Hou, Xiangyu Chen, Jiaqi Bai, Bofan Lyu, Jianfei Yang

机构 * MARS Lab, Nanyang Technological University（MARS实验室，南洋理工大学）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

AI总结提出MARS策略，通过自适应地在需要时引入随机性，在单模态阶段使用确定性学习，平衡生成策略的多模态能力与确定性模型的高效率，在模拟和真实任务中提升成功率和推理速度。

Comments 13 figures, 17 pages

2605.24805 2026-05-26 cs.CV 77%

Fishbone: From One 3D Asset to a Million Controllable Edits

Fishbone: 从一个3D资产到百万可控编辑

Yumeng He, Xiaoying Wang, Peihao Li, Yanjia Huang, Joe Masterjohn, Jiajun Wu, Leonidas Guibas, Yin Yang, Ying Jiang, Chenfanfu Jiang

机构 * UCLA（加州大学洛杉矶分校）； USC（南加州大学）； UC Berkeley（加州大学伯克利分校）； TRI（技术研究院）； Stanford（斯坦福大学）； Utah（犹他大学）

专题命中机器人学习：robotics(abstract)；robot learning(abstract)；embodied AI(abstract)；分类 cs.CV

AI总结提出一种统一的脊-肋表示方法Fishbone，支持通用网格的可控参数化变形、降阶动力学和动画，并构建了Fishbone-136K数据集，应用于可控3D生成、机器人学习数据增强等任务。

Comments 20 pages, 19 figures

2603.25903 2026-04-17 cs.RO 77%

Emergent Neural Automaton Policies: Learning Symbolic Structure from Visuomotor Trajectories

涌现神经自动机策略：从视觉运动轨迹中学习符号结构

Yiyuan Pan, Xusheng Luo, Hanjiang Hu, Peiqi Yu, Changliu Liu

机构 * Robotics Institute, Carnegie Mellon University（卡内基梅隆大学机器人研究所）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

AI总结本文提出ENAP框架，通过从视觉运动轨迹中自适应推断Mealy状态机，结合行为克隆学习连续控制，实现高效且可解释的机器人策略。

2510.00573 2026-04-16 cs.RO 77%

GRITS: A Spillage-Aware Guided Diffusion Policy for Robot Food Scooping Tasks

GRITS：一种考虑泄漏的引导扩散策略用于机器人食物舀取任务

Yen-Ling Tai, Yi-Ru Yang, Kuan-Ting Yu, Yu-Wei Chao, Yi-Ting Chen

机构 * National Yang Ming Chiao Tung University（国立阳明交通大学）； XYZ Robotics（XYZ机器人公司）； NVIDIA

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

AI总结 GRITS通过引导扩散策略减少食物泄漏，提高机器人食物舀取任务的可靠性与成功率，实验表明其在任务成功率和泄漏率上优于基线方法。

2601.16212 2026-03-26 cs.RO 77%

Point Bridge: 3D Representations for Cross Domain Policy Learning

点桥：跨领域策略学习的3D表示

Siddhant Haldar, Lars Johannsmeier, Lerrel Pinto, Abhishek Gupta, Dieter Fox, Yashraj Narang, Ajay Mandlekar

机构 * NVIDIA ； New York University（纽约大学）； University of Washington（华盛顿大学）

专题命中机器人学习：manipulation(abstract)；robot foundation model(abstract)；robotic(abstract)；分类 cs.RO

AI总结点桥通过统一的点表示实现跨领域策略学习，利用视觉语言模型提取点表示，结合Transformer策略学习，无需显式视觉或物体对齐，提升仿真到现实的零样本迁移性能。

2508.18269 2025-10-08 cs.RO 77%

FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Models

Zhide Zhong, Haodong Yan, Junfeng Li, Xiangchen Liu, Xin Gong, Tianran Zhang, Wenxuan Song, Jiayi Chen, Xinhu Zheng, Hesheng Wang, Haoang Li

专题命中机器人学习：robotics(abstract)；manipulation(abstract)；world model(abstract)；分类 cs.RO

2509.25352 2025-10-01 cs.RO 77%

SRMP: Search-Based Robot Motion Planning Library

Itamar Mishani, Yorai Shaoul, Ramkumar Natarajan, Jiaoyang Li, Maxim Likhachev

机构 * Amazon Robotics（亚马逊机器人）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

Comments Submitted for Publication

2409.03685 2025-06-03 cs.RO cs.AI cs.CV cs.LG 77%

View-Invariant Policy Learning via Zero-Shot Novel View Synthesis

Stephen Tian, Blake Wulfe, Kyle Sargent, Katherine Liu, Sergey Zakharov, Vitor Guizilini, Jiajun Wu

机构 * Stanford University（斯坦福大学）； Toyota Research Institute（丰田研究中心）

专题命中机器人学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments Accepted to CoRL 2024

2505.18792 2025-05-27 cs.RO 77%

On the Dual-Use Dilemma in Physical Reasoning and Force

William Xie, Enora Rice, Nikolaus Correll

机构 * University of Colorado Boulder（科罗拉多大学博尔德分校）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robotic(abstract)；分类 cs.RO

2312.04549 2023-12-08 cs.RO cs.AI cs.CV cs.LG cs.SY eess.SY 77%

PlayFusion: Skill Acquisition via Diffusion from Language-Annotated Play

Lili Chen, Shikhar Bahl, Deepak Pathak

专题命中机器人学习：robotics(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments In CoRL 2023. Website at https://play-fusion.github.io

2303.17481 2023-04-05 cs.RO 77%

Teaching contact-rich tasks from visual demonstrations by constraint extraction

Christian Hegeler, Filippo Rozzi, Loris Roveda, Kevin Haninger

专题命中机器人学习：manipulation(abstract)；robot policy(abstract)；robotic(abstract)；分类 cs.RO

Comments Submitted ICRA23. Code available at https://github.com/khaninger/contact_monitoring Video available at https://www.youtube.com/watch?v=Tf-zt8VKu_M

2205.06333 2023-03-14 cs.RO cs.AI cs.CV cs.LG 77%

Visuomotor Control in Multi-Object Scenes Using Object-Aware Representations

Negin Heravi, Ayzaan Wahid, Corey Lynch, Pete Florence, Travis Armstrong, Jonathan Tompson, Pierre Sermanet, Jeannette Bohg, Debidatta Dwibedi

专题命中机器人学习：robotics(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

2207.09450 2022-07-20 cs.RO cs.AI cs.CV cs.LG cs.SY eess.SY 77%

Human-to-Robot Imitation in the Wild

Shikhar Bahl, Abhinav Gupta, Deepak Pathak

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments Published at RSS 2022. Demos at https://human2robot.github.io

1810.04879 2021-11-04 cs.RO cs.AI cs.CV cs.HC cs.LG 77%

Generating Shared Latent Variables for Robots to Imitate Human Movements and Understand their Physical Limitations

Maxime Devanne, Sao Mai Nguyen

专题命中机器人学习：robotics(abstract)；robot learning(abstract)；分类 cs.RO、cs.AI、cs.CV

Journal ref Leal-Taix{é} L; Roth S. Lecture Notes in Computer Science, 11130, Springer, Cham, pp.190-197, 2019, Computer Vision -- ECCV 2018 Workshops

2110.04633 2021-10-12 cs.RO 77%

Credit Assignment Safety Learning from Human Demonstrations

Ahalya Prabhakar, Aude Billard

专题命中机器人学习：robotics(abstract)；robot learning(abstract)；navigation(abstract)；分类 cs.RO

Comments Presented at AI-HRI symposium as part of AAAI-FSS 2021 (arXiv:2109.10836)

1. 机器人学习 858 篇

UniPrototype: Humn-Robot Skill Learning with Uniform Prototypes

VAT: Vision Action Transformer by Unlocking Full Representation of ViT

Learning Generalizable Visuomotor Policy through Dynamics-Alignment

FRMD: Fast Robot Motion Diffusion with Consistency-Distilled Movement Primitives for Smooth Action Generation

Goal-Reaching Policy Learning from Non-Expert Observations via Effective Subgoal Guidance

Learning needle insertion from sample task executions

SAPIEN: A SimulAted Part-based Interactive ENvironment

Learning Deep Neural Network Policies with Continuous Memory States

Neuro-Symbolic Generation of Explanations for Robot Policies with Weighted Signal Temporal Logic

Render and Diffuse: Aligning Image and Action Spaces for Diffusion-based Behaviour Cloning

Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes

Learning Latent Plans from Play

Advances in Multi-agent Reinforcement Learning: Persistent Autonomy and Robot Learning Lab Report 2024

Robot Policy Learning with Temporal Optimal Transport Reward

GeoTLM: Geometry-aware Tactile-Language Models for Contact Motion Orientation Reasoning of Dynamic Objects

MARS Policy: Multimodality Only When It Matters

Fishbone: From One 3D Asset to a Million Controllable Edits

Emergent Neural Automaton Policies: Learning Symbolic Structure from Visuomotor Trajectories

GRITS: A Spillage-Aware Guided Diffusion Policy for Robot Food Scooping Tasks

Point Bridge: 3D Representations for Cross Domain Policy Learning

FlowVLA: Visual Chain of Thought-based Motion Reasoning for Vision-Language-Action Models

SRMP: Search-Based Robot Motion Planning Library

View-Invariant Policy Learning via Zero-Shot Novel View Synthesis

On the Dual-Use Dilemma in Physical Reasoning and Force

PlayFusion: Skill Acquisition via Diffusion from Language-Annotated Play

Teaching contact-rich tasks from visual demonstrations by constraint extraction

Visuomotor Control in Multi-Object Scenes Using Object-Aware Representations

Human-to-Robot Imitation in the Wild

Generating Shared Latent Variables for Robots to Imitate Human Movements and Understand their Physical Limitations

Credit Assignment Safety Learning from Human Demonstrations