机器人 / 具身智能 - arXivDaily 专题

1911.00357 2020-01-22 cs.CV cs.AI cs.LG 75%

DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Frames

Erik Wijmans, Abhishek Kadian, Ari Morcos, Stefan Lee, Irfan Essa, Devi Parikh, Manolis Savva, Dhruv Batra

专题命中模仿学习与强化学习：embodied AI(abstract)；navigation(abstract)；分类 cs.AI、cs.CV、cs.LG

1911.09676 2019-11-22 cs.LG cs.CV cs.RO stat.ML 75%

Third-Person Visual Imitation Learning via Decoupled Hierarchical Controller

Pratyusha Sharma, Deepak Pathak, Abhinav Gupta

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments Accepted at NeurIPS 2019. Videos at https://pathak22.github.io/hierarchical-imitation/

1910.12453 2019-10-29 cs.LG cs.AI cs.RO stat.ML 75%

Asynchronous Methods for Model-Based Reinforcement Learning

Yunzhi Zhang, Ignasi Clavera, Boren Tsai, Pieter Abbeel

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments 10 pages, CoRL 2019

1910.01240 2019-10-04 cs.LG cs.AI cs.RO stat.ML 75%

Deep Reinforcement Learning for Single-Shot Diagnosis and Adaptation in Damaged Robots

Shresth Verma, Haritha S. Nair, Gaurav Agarwal, Joydip Dhar, Anupam Shukla

专题命中模仿学习与强化学习：robotics(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

1903.06282 2019-03-19 cs.RO cs.AI cs.LG 75%

ROS2Learn: a reinforcement learning framework for ROS 2

Yue Leire Erro Nuin, Nestor Gonzalez Lopez, Elias Barba Moral, Lander Usategui San Juan, Alejandro Solano Rueda, Víctor Mayoral Vilches, Risto Kojcev

专题命中模仿学习与强化学习：robotics(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

1903.06278 2019-03-19 cs.RO cs.AI cs.LG 75%

gym-gazebo2, a toolkit for reinforcement learning using ROS 2 and Gazebo

Nestor Gonzalez Lopez, Yue Leire Erro Nuin, Elias Barba Moral, Lander Usategui San Juan, Alejandro Solano Rueda, Víctor Mayoral Vilches, Risto Kojcev

专题命中模仿学习与强化学习：robotics(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

1810.07225 2018-10-18 cs.RO cs.AI cs.LG 75%

Integrating kinematics and environment context into deep inverse reinforcement learning for predicting off-road vehicle trajectories

Yanfu Zhang, Wenshan Wang, Rogerio Bonatti, Daniel Maturana, Sebastian Scherer

专题命中模仿学习与强化学习：robotics(abstract)；navigation(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments CoRL 2018

1802.09564 2018-05-29 cs.RO cs.AI cs.LG 75%

Reinforcement and Imitation Learning for Diverse Visuomotor Skills

Yuke Zhu, Ziyu Wang, Josh Merel, Andrei Rusu, Tom Erez, Serkan Cabi, Saran Tunyasuvunakool, János Kramár, Raia Hadsell, Nando de Freitas, Nicolas Heess

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments 13 pages, 6 figures, Published in RSS 2018

1711.07479 2017-11-22 cs.RO cs.AI cs.LG stat.ML 75%

Teaching a Machine to Read Maps with Deep Reinforcement Learning

Gino Brunner, Oliver Richter, Yuyi Wang, Roger Wattenhofer

专题命中模仿学习与强化学习：robotics(abstract)；navigation(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments Paper accepted at 32nd AAAI Conference on Artificial Intelligence, AAAI 2018, New Orleans, Louisiana, USA

1610.03164 2016-10-12 cs.RO cs.AI cs.CL cs.LG 75%

Navigational Instruction Generation as Inverse Reinforcement Learning with Neural Machine Translation

Andrea F. Daniele, Mohit Bansal, Matthew R. Walter

专题命中模仿学习与强化学习：robotics(abstract)；navigation(abstract)；分类 cs.RO、cs.AI、cs.LG

1603.00448 2016-05-30 cs.LG cs.AI cs.RO 75%

Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization

Chelsea Finn, Sergey Levine, Pieter Abbeel

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments International Conference on Machine Learning (ICML), 2016, to appear

2608.09138 2026-08-12 cs.RO cs.AI 版本更新 74%

SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning

SpeedTuning：用轻量强化学习加速策略执行

David D. Yuan, Tony Z. Zhao, Kaylee Burns, Chelsea Finn

机构 * Stanford University（斯坦福大学）

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI；robotics(journal_ref)

AI总结 SpeedTuning是一种轻量强化学习框架，可预测动作最优执行速度，在无需额外数据采集的情况下，将机器人操作策略加速超2.4倍且保持足够成功率，适用于多种动态精确任务。

Comments 10 pages, 12 figures. This arXiv version includes an appendix with qualitative simulation rollouts and additional ablations. Published at ICRA 2025

Journal ref 2025 IEEE International Conference on Robotics and Automation (ICRA), pp. 1184-1192, 2025

2605.05172 2026-06-17 cs.RO cs.AI 版本更新 74%

When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning

当生活给你行为克隆，就做Q函数：从行为克隆中提取Q值用于机器人强化学习

Lakshita Dodeja, Ondrej Biza, Shivam Vats, Stephen Hart, Stefanie Tellex, Robin Walters, Karl Schmeckpeper, Thomas Weng

机构 * Rai-Inst

专题命中模仿学习与强化学习：robot learning(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI；robotics(comments)

AI总结提出Q2RL算法，通过从行为克隆策略中提取Q函数并利用Q门控切换策略，实现高效的离线到在线强化学习，在机器人操作任务中达到100%成功率和3.75倍提升。

Comments Robotics: Science and Systems, 2026

2506.17639 2026-06-17 cs.RO cs.AI 版本更新 74%

RLRC: Reinforcement Learning-based Recovery for Compressed Vision-Language-Action Models

RLRC：基于强化学习的压缩视觉-语言-动作模型恢复

Yuxuan Chen, Yixin Han, Yize Huang, Xiao Li

机构 * State Key Laboratory of Mechanical System and Vibration（机械系统与振动国家重点实验室）； Shanghai Key Laboratory of Intelligent Robotics（上海智能机器人重点实验室）； School of Mechanical Engineering, Shanghai Jiao Tong University（上海交通大学机械工程学院）

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI；robotics(journal_ref)

AI总结提出RLRC三阶段压缩恢复流程，通过结构化剪枝、SFT和强化学习恢复以及量化，实现8倍内存减少和2.3倍推理加速，同时保持任务成功率。

Comments 8 pages, 10 figures; accepted by RA-L 2026

Journal ref IEEE Robotics and Automation Letters, vol. 11, no. 7, pp. 8864-8871, July 2026

2511.14427 2026-06-11 cs.RO cs.LG 版本更新 74%

Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning

面向接触丰富机器人强化学习的自监督多感官预训练

Rickmer Krohn, Vignesh Prasad, Gabriele Tiboni, Georgia Chalvatzaki

机构 * Interactive Robot Perception & Learning (PEARL) Lab, TU Darmstadt, Germany（图腾机器人感知与学习实验室，图腾施塔德大学，德国）； Hessian.AI（海斯堡人工智能）； Robotics Institute Germany (RIG)（德国机器人研究所（RIG））

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.LG；robotics(journal_ref)

AI总结提出MSDP框架，通过掩码自编码和跨模态预测学习多感官表示，并采用非对称架构（评论家使用交叉注意力提取动态特征，演员使用稳定池化表示）加速策略学习，在模拟和真实机器人任务中展现出鲁棒性和高效性。

Comments 8 pages, 11 figures

Journal ref IEEE Robotics and Automation Letters, 2026, Vol. 11, No. 6, pp. 6799-6806

2601.04686 2026-01-09 cs.LG cs.RO 74%

Nightmare Dreamer: Dreaming About Unsafe States And Planning Ahead

噩梦梦游者：对不安全状态的梦境与提前规划

Oluwatosin Oseni, Shengjie Wang, Jun Zhu, Micah Corah

专题命中模仿学习与强化学习：robotics(abstract,comments)；world model(abstract)；分类 cs.RO、cs.LG

AI总结 Nightmare Dreamer是一种基于模型的安全强化学习算法，通过预测潜在安全违规并规划动作，在机器人控制任务中实现了接近零安全违规和高奖励效率。

Comments RSS'25: Multi-Objective Optimization and Planning in Robotics Workshop: 5 pages, 8 figures

2512.17180 2025-12-24 cs.RO cs.AI 74%

Conservative Bias in Multi-Teacher Learning: Why Agents Prefer Low-Reward Advisors

多教师学习中的保守偏见：为什么智能体更偏好低奖励顾问

Maher Mesto, Francisco Cruz

机构 * School of Computer Science and Engineering, University of New South Wales（新南威尔士大学计算机科学与工程学院）； Escuela de Ingeniería, Universidad Central de Chile（智利中央大学工程学院）

专题命中模仿学习与强化学习：navigation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI；robotics(comments)

AI总结本文揭示了多教师学习中智能体偏好保守低奖励顾问的现象，发现其受保守偏见主导，并在特定阈值下失效，同时在概念漂移情况下显著优于基线Q学习。

Comments 10 pages, 5 figures. Accepted at ACRA 2025 (Australasian Conference on Robotics and Automation)

2509.12507 2025-09-17 cs.RO cs.HC cs.LG 74%

Learning to Generate Pointing Gestures in Situated Embodied Conversational Agents

Anna Deichler, Siyang Wang, Simon Alexanderson, Jonas Beskow

机构 * Division of Speech, Music and Hearing, KTH Royal Institute of Technology（语音、音乐与听觉系，皇家理工学院）

专题命中模仿学习与强化学习：robotics(abstract,journal_ref)；embodied agent(abstract)；分类 cs.RO、cs.LG

Comments DOI: 10.3389/frobt.2023.1110534. This is the author's LaTeX version

Journal ref Frontiers in Robotics and AI, 10:1110534 (2023)

2403.10794 2025-05-12 cs.RO cs.LG cs.MA 74%

Diffusion-Reinforcement Learning Hierarchical Motion Planning in Multi-agent Adversarial Games

Zixuan Wu, Sean Ye, Manisha Natarajan, Matthew C. Gombolay

机构 * Institute of Robotics and Intelligent Machines (IRIM)（机器人与智能机械研究所）； Georgia Institute of Technology（佐治亚理工学院）

专题命中模仿学习与强化学习：manipulation(abstract)；navigation(abstract)；分类 cs.RO、cs.LG；robotics(comments)

Comments This work has been submitted to the IEEE Robotics and Automation Letters (RA-L) for possible publication

2409.09990 2025-04-29 cs.LG cs.NE cs.RO 74%

SHIRE: Enhancing Sample Efficiency using Human Intuition in REinforcement Learning

Amogh Joshi, Adarsh Kumar Kosta, Kaushik Roy

机构 * Purdue University（普渡大学）

专题命中模仿学习与强化学习：robotics(abstract,comments)；robotic(abstract)；分类 cs.RO、cs.LG

Comments Accepted for publication at the IEEE International Conference on Robotics & Automation (ICRA) 2025

2504.10002 2025-04-15 cs.RO cs.LG 74%

FLoRA: Sample-Efficient Preference-based RL via Low-Rank Style Adaptation of Reward Functions

Daniel Marta, Simon Holk, Miguel Vasco, Jens Lundell, Timon Homberger, Finn Busch, Olov Andersson, Danica Kragic, Iolanda Leite

专题命中模仿学习与强化学习：robotics(abstract,comments)；robotic(abstract)；分类 cs.RO、cs.LG

Comments Accepted at 2025 IEEE International Conference on Robotics & Automation (ICRA). We provide videos of our results and source code at https://sites.google.com/view/preflora/

2411.11406 2024-11-19 cs.LG cs.RO 74%

Bridging the Resource Gap: Deploying Advanced Imitation Learning Models onto Affordable Embedded Platforms

Haizhou Ge, Ruixiang Wang, Zhu-ang Xu, Hongrui Zhu, Ruichen Deng, Yuhang Dong, Zeyu Pang, Guyue Zhou, Junyu Zhang, Lu Shi

专题命中模仿学习与强化学习：robotics(abstract,comments)；manipulation(abstract)；分类 cs.RO、cs.LG

Comments Accepted by the 2024 IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO 2024)

2410.18800 2024-10-25 cs.LG cs.RO 74%

PointPatchRL -- Masked Reconstruction Improves Reinforcement Learning on Point Clouds

Balázs Gyenes, Nikolai Franke, Philipp Becker, Gerhard Neumann

专题命中模仿学习与强化学习：robotics(abstract)；manipulation(abstract)；分类 cs.RO、cs.LG；robot learning(comments)

Comments 18 pages, 15 figures, accepted for publication at the 8th Conference on Robot Learning (CoRL 2024)

2312.00344 2023-12-04 cs.RO cs.LG 74%

TRC: Trust Region Conditional Value at Risk for Safe Reinforcement Learning

Dohyeong Kim, Songhwai Oh

专题命中模仿学习与强化学习：robotics(abstract,journal_ref)；navigation(abstract)；分类 cs.RO、cs.LG

Comments RA-L and ICRA 2022

Journal ref IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 2621-2628, April 2022

2209.13052 2023-05-04 cs.RO cs.AI 74%

Training Efficient Controllers via Analytic Policy Gradient

Nina Wiedemann, Valentin Wüest, Antonio Loquercio, Matthias Müller, Dario Floreano, Davide Scaramuzza

专题命中模仿学习与强化学习：robotics(abstract,journal_ref)；robotic(abstract)；分类 cs.RO、cs.AI

Journal ref IEEE Conference on Robotics and Automation (ICRA 2023)

2303.09628 2023-03-20 cs.LG cs.RO 74%

Efficient Learning of High Level Plans from Play

Núria Armengol Urpí, Marco Bagatella, Otmar Hilliges, Georg Martius, Stelian Coros

专题命中模仿学习与强化学习：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.LG；robotics(comments)

Comments Accepted to the International Conference on Robotics and Automation 2023

2205.10431 2022-05-30 cs.LG cs.AI 74%

Learning Dense Reward with Temporal Variant Self-Supervision

Yuning Wu, Jieliang Luo, Hui Li

专题命中模仿学习与强化学习：manipulation(abstract,comments)；robotic(abstract)；分类 cs.AI、cs.LG

Comments 4 pages, 6 figures, accepted to ICRA 2022 RL for Contact-Rich Manipulation Workshop

2112.12288 2022-01-25 cs.LG cs.RO cs.SY eess.SY 74%

Safety and Liveness Guarantees through Reach-Avoid Reinforcement Learning

Kai-Chieh Hsu, Vicenç Rubies-Royo, Claire J. Tomlin, Jaime F. Fisac

专题命中模仿学习与强化学习：robotics(abstract,comments)；robotic(abstract)；分类 cs.RO、cs.LG

Comments Accepted in Robotics: Science and Systems (RSS), 2021

2107.08325 2021-07-20 cs.RO cs.AI cs.SY eess.SY 74%

Vision-Based Autonomous Car Racing Using Deep Imitative Reinforcement Learning

Peide Cai, Hengli Wang, Huaiyang Huang, Yuxuan Liu, Ming Liu

专题命中模仿学习与强化学习：world model(abstract)；robotic(abstract)；分类 cs.RO、cs.AI；robotics(comments)

Comments 8 pages, 8 figures. IEEE Robotics and Automation Letters (RA-L) & IROS 2021

2104.02863 2021-04-08 cs.RO cs.LG 74%

The Value of Planning for Infinite-Horizon Model Predictive Control

Nathan Hatch, Byron Boots

专题命中模仿学习与强化学习：robotics(abstract,comments)；navigation(abstract)；分类 cs.RO、cs.LG

Comments 7 pages, 8 figures. To appear in the proceedings of the International Conference on Robotics and Automation (ICRA) 2021

1. 模仿学习与强化学习 4111 篇

DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Frames

Third-Person Visual Imitation Learning via Decoupled Hierarchical Controller

Asynchronous Methods for Model-Based Reinforcement Learning

Deep Reinforcement Learning for Single-Shot Diagnosis and Adaptation in Damaged Robots

ROS2Learn: a reinforcement learning framework for ROS 2

gym-gazebo2, a toolkit for reinforcement learning using ROS 2 and Gazebo

Integrating kinematics and environment context into deep inverse reinforcement learning for predicting off-road vehicle trajectories

Reinforcement and Imitation Learning for Diverse Visuomotor Skills

Teaching a Machine to Read Maps with Deep Reinforcement Learning

Navigational Instruction Generation as Inverse Reinforcement Learning with Neural Machine Translation

Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization

SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning

When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning

RLRC: Reinforcement Learning-based Recovery for Compressed Vision-Language-Action Models

Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning

Nightmare Dreamer: Dreaming About Unsafe States And Planning Ahead

Conservative Bias in Multi-Teacher Learning: Why Agents Prefer Low-Reward Advisors

Learning to Generate Pointing Gestures in Situated Embodied Conversational Agents

Diffusion-Reinforcement Learning Hierarchical Motion Planning in Multi-agent Adversarial Games

SHIRE: Enhancing Sample Efficiency using Human Intuition in REinforcement Learning

FLoRA: Sample-Efficient Preference-based RL via Low-Rank Style Adaptation of Reward Functions

Bridging the Resource Gap: Deploying Advanced Imitation Learning Models onto Affordable Embedded Platforms

PointPatchRL -- Masked Reconstruction Improves Reinforcement Learning on Point Clouds

TRC: Trust Region Conditional Value at Risk for Safe Reinforcement Learning

Training Efficient Controllers via Analytic Policy Gradient

Efficient Learning of High Level Plans from Play

Learning Dense Reward with Temporal Variant Self-Supervision

Safety and Liveness Guarantees through Reach-Avoid Reinforcement Learning

Vision-Based Autonomous Car Racing Using Deep Imitative Reinforcement Learning

The Value of Planning for Infinite-Horizon Model Predictive Control