机器人 / 具身智能 - arXivDaily 专题

2105.04607 2021-06-02 cs.RO cs.AI cs.LG 82%

Efficient Self-Supervised Data Collection for Offline Robot Learning

Shadi Endrawis, Gal Leibovich, Guy Jacob, Gal Novik, Aviv Tamar

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

Comments Accepted in ICRA 2021

2103.14633 2021-03-29 cs.RO cs.CV cs.LG cs.NE 82%

Visionary: Vision architecture discovery for robot learning

Iretiayo Akinola, Anelia Angelova, Yao Lu, Yevgen Chebotar, Dmitry Kalashnikov, Jacob Varley, Julian Ibarz, Michael S. Ryoo

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

Journal ref ICRA 2021

2004.10301 2020-04-23 cs.RO cs.AI cs.LG cs.SY eess.SY 82%

Structured Mechanical Models for Robot Learning and Control

Jayesh K. Gupta, Kunal Menda, Zachary Manchester, Mykel J. Kochenderfer

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

Comments First two authors contributed equally. Accepted at L4DC2020. Source code and videos at https://sites.google.com/stanford.edu/smm/

2001.07343 2020-01-22 cs.RO cs.AI cs.LG cs.SY eess.SY 82%

Lyceum: An efficient and scalable ecosystem for robot learning

Colin Summers, Kendall Lowrey, Aravind Rajeswaran, Siddhartha Srinivasa, Emanuel Todorov

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

1906.05930 2019-09-24 cs.CV cs.LG cs.RO 82%

Cross-View Policy Learning for Street Navigation

Ang Li, Huiyi Hu, Piotr Mirowski, Mehrdad Farajtabar

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

Comments 8 pages

1905.11602 2019-06-05 cs.LG cs.AI cs.RO stat.ML 82%

Differentiable Algorithm Networks for Composable Robot Learning

Peter Karkus, Xiao Ma, David Hsu, Leslie Pack Kaelbling, Wee Sun Lee, Tomas Lozano-Perez

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

Comments RSS 2019 camera ready. Video is available at https://youtu.be/4jcYlTSJF4Y

1804.07269 2019-02-18 cs.RO cs.AI cs.LG 82%

Socially Guided Intrinsic Motivation for Robot Learning of Motor Skills

Sao Mai Nguyen, Pierre-Yves Oudeyer

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

Journal ref Autonomous Robots, Springer Verlag, 2014, 36 (3), pp.273-294

1710.06542 2017-10-19 cs.RO cs.AI cs.LG 82%

Asymmetric Actor Critic for Image-Based Robot Learning

Lerrel Pinto, Marcin Andrychowicz, Peter Welinder, Wojciech Zaremba, Pieter Abbeel

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

Comments Videos of experiments can be found at http://www.goo.gl/b57WTs

1605.05045 2017-03-01 stat.ML cs.CV cs.LG cs.RO 82%

Incremental Robot Learning of New Objects with Fixed Update Time

Raffaello Camoriano, Giulia Pasquale, Carlo Ciliberto, Lorenzo Natale, Lorenzo Rosasco, Giorgio Metta

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

Comments 8 pages, 3 figures

1701.01497 2017-01-09 cs.AI cs.LG cs.RO 82%

Learning local trajectories for high precision robotic tasks : application to KUKA LBR iiwa Cartesian positioning

Joris Guerin, Olivier Gibaru, Eric Nyiri, Stephane Thiery

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

Comments 6 pages, double column, 6 figures and one table. Published in: Industrial Electronics Society , IECON 2016 - 42nd Annual Conference of the IEEE

Journal ref Industrial Electronics Society, IECON 2016-42nd Annual Conference of the IEEE Pages 5316--5321

2408.02297 2025-01-15 cs.RO cs.CV 82%

Perception Matters: Enhancing Embodied AI with Uncertainty-Aware Semantic Segmentation

Sai Prasanna, Daniel Honerkamp, Kshitij Sirohi, Tim Welschehold, Wolfram Burgard, Abhinav Valada

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

Journal ref Proceedings of the International Symposium on Robotics Research (ISRR), 2024

2202.06843 2023-04-14 cs.RO cs.LG 82%

Continual Learning from Demonstration of Robotics Skills

Sayantan Auddy, Jakob Hollenstein, Matteo Saveriano, Antonio Rodríguez-Sánchez, Justus Piater

专题命中机器人学习：robotics(title,comments)；robotic(abstract)；分类 cs.RO、cs.LG

Comments To appear in Robotics and Autonomous Systems

2111.00956 2022-01-19 cs.RO cs.LG 82%

Robot Learning from Randomized Simulations: A Review

Fabio Muratore, Fabio Ramos, Greg Turk, Wenhao Yu, Michael Gienger, Jan Peters

专题命中机器人学习：robot learning(title)；robotics(abstract,comments)；分类 cs.RO、cs.LG

Comments submitted to Frontiers in Robotics and AI

1912.05629 2019-12-13 cs.LG cs.RO stat.ML 82%

Large-scale Kernel Methods and Applications to Lifelong Robot Learning

Raffaello Camoriano

专题命中机器人学习：robot learning(title)；robotics(abstract,comments)；分类 cs.RO、cs.LG

Comments Ph. D. Thesis for the Doctoral Course in Bioengineering and Robotics (Curriculum in Humanoid Robotics) at Università degli Studi di Genova, in collaboration with Istituto Italiano di Tecnologia. Advisors: Prof. Giorgio Metta and Prof. Lorenzo Rosasco

2607.24296 2026-07-28 cs.RO 新提交 81%

PAC-DP: PAC-Bayesian Diffusion Policy Learning

PAC-DP：PAC贝叶斯扩散策略学习

Mohammad Hasan Yeganegi, Dian Yu, Andrea Del Prete, Majid Khadiv, Matteo Saveriano

机构 * University of Trento（特伦托大学）； Munich Institute of Robotics and Machine Intelligence (MIRMI), Technical University of Munich (TUM)（慕尼黑工业大学慕尼黑机器人与机器智能研究所）

专题命中机器人学习：robotics(abstract)；manipulation(abstract)；robot policy(abstract)；robotic(abstract)

AI总结研究针对扩散策略在有限数据下泛化控制有限的问题，提出PAC-DP，将DP建模为贝叶斯神经网络并定义泛化界导出新训练目标。该方法理论上规范DP训练，实验显示在多基准测试中性能提升，尤其在低数据和复杂任务中效果显著。

2511.03882 2026-05-25 cs.CV cs.AI cs.LG cs.RO 81%

Investigating Robot Control Policy Learning for Autonomous X-ray-guided Spine Procedures

自主X光引导脊柱手术的机器人控制策略学习研究

Florence Klitzner, Blanca Inigo, Benjamin D. Killeen, Lalithkumar Seenivasan, Michelle Song, Axel Krieger, Mathias Unberath

机构 * Johns Hopkins University（约翰霍普金斯大学）； Technical University of Munich（慕尼黑技术大学）； Johns Hopkins School of Medicine（约翰霍普金斯医学院）

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

AI总结本文研究基于模仿学习的机器人控制策略在稀疏输入（双平面X光）下进行椎体成形术中的可行性，通过模拟环境训练策略实现基于视觉的套管针迭代对齐，首次尝试成功率达68.5%，并展示了向复杂解剖结构和真实X光的迁移能力，但存在入口点精度不足等局限。

2511.07732 2026-03-31 cs.RO cs.AI cs.CL cs.CV cs.LG 81%

ViPRA: Video Prediction for Robot Actions

ViPRA：用于机器人动作的视频预测

Sandeep Routray, Hengkai Pan, Unnat Jain, Shikhar Bahl, Deepak Pathak

机构 * Carnegie Mellon University（卡内基梅隆大学）； Skild AI ； University of California, Irvine（加州大学尔湾分校）

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

AI总结 ViPRA通过预训练和微调框架，从无标注视频中学习连续机器人控制，利用视频语言模型预测视觉观察和运动中心潜在动作，支持跨机器人形态的泛化和高频率连续控制。

Comments In ICLR 2026. Website: https://vipra-project.github.io

2603.10448 2026-03-24 cs.RO 81%

DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control

DiT4DiT：联合建模视频动态与动作以实现通用机器人控制

Teli Ma, Jia Zheng, Zifan Wang, Chunli Jiang, Andy Cui, Junwei Liang, Shuo Yang

机构 * Mondo Robotics（Mondo机器人公司）； HKUST(GZ)（香港科技大学（广州））； HKUST（香港科技大学）

专题命中机器人学习：robot learning(abstract)；manipulation(abstract)；robot policy(abstract)；robotic(abstract)

AI总结本文提出DiT4DiT模型，通过结合视频扩散变换器与动作扩散变换器，实现视频动态与动作的联合建模，提升机器人控制的泛化能力与样本效率。

Comments https://dit4dit.github.io/

2502.07282 2025-11-18 cs.RO 81%

Leader-follower formation enabled by pressure sensing in free-swimming undulatory robotic fish

Kundan Panta, Hankun Deng, Micah DeLattre, Bo Cheng

机构 * Department of Mechanical Engineering, The Pennsylvania State University（机械工程系，宾夕法尼亚州立大学）

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

Comments 7 pages, 10 figures. Accepted for 2025 IEEE International Conference on Robotics and Automation (ICRA). Supplementary video: https://youtu.be/DIDYGi9Td0I

Journal ref 2025 IEEE International Conference on Robotics and Automation (ICRA), Atlanta, GA, USA, 2025, pp. 14685-14691

2406.07398 2024-08-12 cs.CV cs.AI cs.LG cs.RO 81%

Visual Representation Learning with Stochastic Frame Prediction

Huiwon Jang, Dongyoung Kim, Junsu Kim, Jinwoo Shin, Pieter Abbeel, Younggyo Seo

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

Comments International Conference on Machine Learning (ICML) 2024

2107.00339 2021-07-02 cs.RO cs.AI cs.LG 81%

Policy Transfer across Visual and Dynamics Domain Gaps via Iterative Grounding

Grace Zhang, Linghan Zhong, Youngwoon Lee, Joseph J. Lim

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

Comments Robotics: Science and Systems (RSS), 2021

2608.16978 2026-08-19 cs.RO cs.LG 新提交 81%

VLCP: Vision Language Control Policy Closed-Loop Code Replanning for Robot Manipulation

VLCP：用于机器人操纵的视觉语言控制策略闭环代码重规划

Dhia Naouali, Minghan Wu, Claudia Wong, Abhinav Puthran, Omar G. Younis

机构 * University of Monastir（莫纳斯提尔大学）； St. Mildred’s-Lightbourn School（圣米尔德雷德-莱特本学校）； Cornell University（康奈尔大学）； Carnegie Mellon University（卡内基梅隆大学）； Silverstream AI（银流人工智能公司）； Mila -- Quebec AI Institute（米拉-魁北克人工智能研究所）

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

AI总结 VLCP是一种无需训练的机器人操纵策略，通过在单回合内每K步由VLM重写控制代码闭合循环，在57项任务的评估中综合成功率达35.1%，较开环策略提升十倍。

2608.15002 2026-08-18 cs.RO cs.AR cs.LG 新提交 81%

NPU Offloading of a Frozen Visual Encoder for Robot Policy Training

用于机器人策略训练的冻结视觉编码器的NPU卸载

Hyojun Yun, Seungjae Won, Hyungpil Moon

机构 * Sungkyunkwan University（成均馆大学）

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

AI总结本文探究冻结视觉编码器的NPU卸载对机器人策略训练的影响，构建异步训练流水线，实验发现该方法可降低训练能耗，但会增加训练时间、降低策略成功率。

Comments 6 pages, 4 figures, 4 tables

2607.24159 2026-07-28 cs.RO cs.CV 新提交 81%

DeVA: Decoupled Video-Action Model with physical guidance for robot policy learning

DeVA：用于机器人策略学习的具有物理引导的解耦视频-动作模型

Mengqi Zhang, Sahil Khose, Simar Kareer, Yuchen Song, Unnat Jain, Judy Hoffman

机构 * University of California, Irvine（加利福尼亚大学欧文分校）； Georgia Institute of Technology（佐治亚理工学院）

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

AI总结研究可泛化机器人操纵策略，提出DeVA模型，通过专门的视频和动作专家、多级特征转移及物理显著引导，实现丰富信息交换，使策略学习更易处理，在模拟和实际实验中展现良好性能。

Comments Project page with videos, code, and checkpoints: https://deva-model.github.io/

2607.22434 2026-07-27 cs.RO cs.AI 新提交 81%

Robot Learning to Communicate through Projected Visual Abstractions

机器人通过投影视觉抽象进行通信学习

Danyang Yan, Boyuan Wang, Jiaxun Liu, Boyuan Chen

机构 * Duke University（杜克大学）

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

AI总结研究让机器人通过投影视觉抽象通信，提出含21自由度灵巧手和影子自我模型的系统，能动态表达影子。通过自我探索学习映射，依目标优化配置，经模拟获可行运动，还引入多种优化，在多场景展示影子表达，建立相关框架。

Comments Our project website is at:https://generalroboticslab.com/shadow

2606.14409 2026-07-21 cs.RO cs.AI 版本更新 81%

Hy-Embodied-0.5-VLA: From Vision-Language-Action Models to a Real-World Robot Learning Stack

Hy-Embodied-0.5-VLA：从视觉-语言-动作模型到真实世界机器人学习栈

He Zhang, Lingzhu Xiang, Haitao Lin, Zeyu Huang, Minghui Wang, Dingyan Zhong, Yubo Dong, Yihao Wu, Yongming Rao, Dongsheng Zhang, Wanjia He, Ling Chen, Kai Huang, Jiahao Chen, Sichang Su, Xumin Yu, Ziyi Wang, Chengwei Zhu, Xiao Teng, Yuchun Guo, Yufeng Zhang, Yuandong Liu, Rui Wang, Zisheng Lu, Han Hu, Zhengyou Zhang

机构 * University of Science and Technology of China（中国科学技术大学）； Tsinghua University（清华大学）

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

AI总结提出端到端机器人学习栈HyVLA-0.5，涵盖数据收集、模型设计、预训练与微调、RL后训练及真实部署，各组件协同工作。

2606.30101 2026-06-30 cs.RO cs.CV 81%

SIR: Structured Image Representations for Explainable Robot Learning

SIR: 用于可解释机器人学习的结构化图像表示

Paul Mattes, Jan Schwab, Jens Bosch, Nils Blank, Maximilian Xiling Li, Minh-Trung Tang, Moritz Haberland, Rudolf Lioutikov

机构 * Intuitive Robots Lab, Karlsruhe Institute of Technology, Germany（直观机器人实验室，卡尔斯鲁厄理工学院，德国）； Robotics Institute Germany（德国机器人研究所）

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

AI总结提出结构化图像表示（SIR），利用场景图作为中间表示，通过端到端稀疏化图结构学习任务相关子图，实现内在可解释性，在RoboCasa上平均成功率19.5%优于基线14.81%，并揭示数据集偏差。

Comments Published at CVPR 2026

Journal ref In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2026. S. 42484-42493

2505.17006 2026-06-19 cs.CV cs.RO 版本更新 81%

CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning

CoMo: 从互联网视频中学习连续潜在运动以实现可扩展的机器人学习

Jiange Yang, Yansong Shi, Haoyi Zhu, Mingyu Liu, Kaijing Ma, Yating Wang, Gangshan Wu, Tong He, Limin Wang

机构 * Nanjing University（南京大学）； Shanghai AI Lab（上海人工智能实验室）； University of Science and Technology of China（中国科学技术大学）； Zhejiang University（浙江大学）； Fudan University（复旦大学）； Tongji University（同济大学）

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

AI总结提出CoMo方法，通过早期时间差分和时序对比学习从互联网视频中学习连续潜在运动，避免离散化信息损失，实现零样本泛化生成伪动作标签，联合训练策略在仿真和真实实验中表现优异。

Comments CVPR 2026

2506.05985 2026-05-29 cs.LG cs.RO 81%

Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning

动态渐进式参数高效专家库混合用于终身机器人学习

Yuheng Lei, Sitong Mao, Shunbo Zhou, Hongyuan Zhang, Xuelong Li, Ping Luo

机构 * The University of Hong Kong（香港大学）； Institute of Artificial Intelligence (TeleAI), China Telecom（人工智能研究院（TeleAI），中国电信）； Huawei Cloud Computing Technologies（华为云计算技术）； Ola Dimensions ； HKU Shanghai Intelligent Computing Research Center（香港大学上海智能计算研究中心）

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

AI总结针对终身学习中任务标识不可用和知识隔离问题，提出动态渐进式参数高效专家库混合（DMPEL），通过构建低秩专家库和轻量路由器实现灵活的前向迁移，并引入专家系数回放缓解遗忘，在LIBERO基准上以最少可训练参数和存储超越现有方法。

Comments Accepted to Transactions on Machine Learning Research (TMLR) at https://openreview.net/forum?id=MHVBrjS8cG . Code is available at https://github.com/HarryLui98/DMPEL

2510.15510 2026-04-09 cs.CV cs.RO 81%

Exploring Conditions for Diffusion models in Robotic Control

探索扩散模型在机器人控制中的条件

Heeseong Shin, Byeongho Heo, Dongyoon Han, Seungryong Kim, Taekyung Kim

机构 * KAIST AI（韩国科学技术院人工智能学院）； NAVER AI Lab（NAVER人工智能实验室）

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

AI总结本文研究利用预训练文本到图像扩散模型获取任务适应的视觉表示，提出ORCA方法以动态视觉信息提升机器人控制性能。

Comments Accepted to CVPR 2026. Project page: https://orca-rc.github.io/

1. 机器人学习 867 篇

Efficient Self-Supervised Data Collection for Offline Robot Learning

Visionary: Vision architecture discovery for robot learning

Structured Mechanical Models for Robot Learning and Control

Lyceum: An efficient and scalable ecosystem for robot learning

Cross-View Policy Learning for Street Navigation

Differentiable Algorithm Networks for Composable Robot Learning

Socially Guided Intrinsic Motivation for Robot Learning of Motor Skills

Asymmetric Actor Critic for Image-Based Robot Learning

Incremental Robot Learning of New Objects with Fixed Update Time

Learning local trajectories for high precision robotic tasks : application to KUKA LBR iiwa Cartesian positioning

Perception Matters: Enhancing Embodied AI with Uncertainty-Aware Semantic Segmentation

Continual Learning from Demonstration of Robotics Skills

Robot Learning from Randomized Simulations: A Review

Large-scale Kernel Methods and Applications to Lifelong Robot Learning

PAC-DP: PAC-Bayesian Diffusion Policy Learning

Investigating Robot Control Policy Learning for Autonomous X-ray-guided Spine Procedures

ViPRA: Video Prediction for Robot Actions

DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control

Leader-follower formation enabled by pressure sensing in free-swimming undulatory robotic fish

Visual Representation Learning with Stochastic Frame Prediction

Policy Transfer across Visual and Dynamics Domain Gaps via Iterative Grounding

VLCP: Vision Language Control Policy Closed-Loop Code Replanning for Robot Manipulation

NPU Offloading of a Frozen Visual Encoder for Robot Policy Training

DeVA: Decoupled Video-Action Model with physical guidance for robot policy learning

Robot Learning to Communicate through Projected Visual Abstractions

Hy-Embodied-0.5-VLA: From Vision-Language-Action Models to a Real-World Robot Learning Stack

SIR: Structured Image Representations for Explainable Robot Learning

CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning

Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning

Exploring Conditions for Diffusion models in Robotic Control