机器人 / 具身智能 - arXivDaily 专题

1709.03339 2018-02-28 cs.AI cs.RO 62%

Autonomous Quadrotor Landing using Deep Reinforcement Learning

Riccardo Polvara, Massimiliano Patacchiola, Sanjay Sharma, Jian Wan, Andrew Manning, Robert Sutton, Angelo Cangelosi

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

Comments The actual copy of the manuscript is a revised version resubmitted to IROS

1709.07643 2018-02-26 cs.RO cs.AI 62%

OptLayer - Practical Constrained Optimization for Deep Reinforcement Learning in the Real World

Tu-Hoa Pham, Giovanni De Magistris, Ryuki Tachibana

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

Comments To appear at ICRA 2018. Video: https://www.youtube.com/watch?v=7liBbk3VjWQ

1704.06676 2018-02-26 cs.AI cs.RO 62%

Modular Multi-Objective Deep Reinforcement Learning with Decision Values

Tomasz Tajmajer

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

1610.04213 2017-12-13 cs.RO cs.AI 62%

Reset-free Trial-and-Error Learning for Robot Damage Recovery

Konstantinos Chatzilygeroudis, Vassilis Vassiliades, Jean-Baptiste Mouret

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

Comments 18 pages, 16 figures, 3 tables, 6 pseudocodes/algorithms, video at https://youtu.be/IqtyHFrb3BU, code at https://github.com/resibots/chatzilygeroudis_2018_rte

1707.06658 2017-11-30 cs.LG cs.AI 62%

RAIL: Risk-Averse Imitation Learning

Anirban Santara, Abhishek Naik, Balaraman Ravindran, Dipankar Das, Dheevatsa Mudigere, Sasikanth Avancha, Bharat Kaul

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

Comments Accepted for presentation in Deep Reinforcement Learning Symposium at NIPS 2017

1711.06782 2017-11-21 cs.LG cs.RO 62%

Leave no Trace: Learning to Reset for Safe and Autonomous Reinforcement Learning

Benjamin Eysenbach, Shixiang Gu, Julian Ibarz, Sergey Levine

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

Comments Videos of our experiments are available at: https://sites.google.com/site/mlleavenotrace/

1709.08233 2017-09-26 cs.AI cs.RO 62%

Learning Unmanned Aerial Vehicle Control for Autonomous Target Following

Siyi Li, Tianbo Liu, Chi Zhang, Dit-Yan Yeung, Shaojie Shen

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

1709.08201 2017-09-26 cs.AI cs.LG stat.ML 62%

An Optimal Online Method of Selecting Source Policies for Reinforcement Learning

Siyuan Li, Chongjie Zhang

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

1709.06166 2017-09-20 cs.AI cs.RO 62%

DropoutDAgger: A Bayesian Approach to Safe Imitation Learning

Kunal Menda, Katherine Driggs-Campbell, Mykel J. Kochenderfer

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

1702.08074 2017-04-26 cs.LG cs.RO 62%

Learning Control for Air Hockey Striking using Deep Reinforcement Learning

Ayal Taitler, Nahum Shimkin

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

Comments Corrected typos Graphs added in results section

1703.01274 2017-03-16 cs.AI cs.HC cs.RO 62%

Actor-Critic Reinforcement Learning with Simultaneous Human Control and Feedback

Kory W. Mathewson, Patrick M. Pilarski

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

1703.00472 2017-03-03 cs.RO cs.LG 62%

Reinforcement Learning for Pivoting Task

Rika Antonova, Silvia Cruciani, Christian Smith, Danica Kragic

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

Comments (Rika Antonova and Silvia Cruciani contributed equally)

1612.03471 2017-03-03 cs.AI cs.RO 62%

Reinforcement Learning With Temporal Logic Rewards

Xiao Li, Cristian-Ioan Vasile, Calin Belta

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

1701.04143 2017-01-17 cs.LG cs.AI 62%

Vulnerability of Deep Reinforcement Learning to Policy Induction Attacks

Vahid Behzadan, Arslan Munir

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

Comments 14 pages, 5 figures, pre-print of submission to MLDM '17

1611.02779 2016-11-11 cs.AI cs.LG cs.NE stat.ML 62%

RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning

Yan Duan, John Schulman, Xi Chen, Peter L. Bartlett, Ilya Sutskever, Pieter Abbeel

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

Comments 14 pages. Under review as a conference paper at ICLR 2017

1608.06235 2016-09-13 cs.RO cs.LG 62%

Adaptive Probabilistic Trajectory Optimization via Efficient Approximate Inference

Yunpeng Pan, Xinyan Yan, Evangelos Theodorou, Byron Boots

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

1606.01178 2016-06-06 cs.CV cs.RO 62%

Reinforcement Learning for Semantic Segmentation in Indoor Scenes

Md. Alimoor Reza, Jana Kosecka

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

1509.06791 2016-02-17 cs.LG cs.RO 62%

Learning Deep Control Policies for Autonomous Aerial Vehicles with MPC-Guided Policy Search

Tianhao Zhang, Gregory Kahn, Sergey Levine, Pieter Abbeel

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

1105.1749 2015-03-18 cs.AI cs.RO cs.SE 62%

A Real-Time Model-Based Reinforcement Learning Architecture for Robot Control

Todd Hester, Michael Quinlan, Peter Stone

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

Comments Added a reference Presents a real-time parallel architecture for model-based reinforcement learning methods

1204.2235 2012-04-11 cs.RO cs.AI cs.DL 62%

Publishing Identifiable Experiment Code And Configuration Is Important, Good and Easy

Richard Vaughan, Jens Wawerla

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

Comments 11 pages

2506.07223 2026-08-11 cs.AI 版本更新 61%

Reflex First, Reflect Later: Latency-Aware Embodied LLM Agents for Dynamic Response

先反射，后反思：面向动态响应的延迟感知具身大语言模型智能体

Yangqing Zheng, Shunqi Mao, Dingxin Zhang, Weidong Cai

机构 * School of Computer Science, The University of Sydney（计算机科学学院，悉尼大学）

专题命中模仿学习与强化学习：embodied agent(abstract)；分类 cs.AI；embodied AI(comments)

AI总结该研究针对动态环境中具身LLM智能体的推理延迟问题，提出RRARA智能体及相关评估指标，通过时间转换机制与预规划器实现决策质量与响应能力的平衡。

Comments Accepted by the CVPR 2025 Embodied AI Workshop

2607.27494 2026-07-31 cs.RO 新提交 61%

Simulation of Surgical Suturing Using Position-Based Dynamics and the Material Point Method for Robot Reinforcement Learning

基于位置动力学（PBD）和物质点法（MPM）的手术缝合模拟用于机器人强化学习

Tleukhan Mussin, Yafei Ou, Mahdi Tavakoli

机构 * University of Alberta（阿尔伯塔大学）

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

AI总结该研究提出基于PBD与MPM的缝合模拟环境，优化GPU执行并构建RL缝合子任务环境，训练的RL智能体在进针、拔针任务中分别达到80%、68%的成功率。

Comments 7 pages, 9 figures, accepted for the IEEE RAS/EMBS 11th International Conference on Biomedical Robotics and Biomechatronics (BioRob 2026)

2602.05608 2026-06-16 cs.RO 版本更新 61%

HiCrowd: Hierarchical Crowd Flow Alignment for Dense Human Environments

HiCrowd：密集人群环境中的分层人群流对齐

Yufei Zhu, Shih-Min Yang, Martin Magnusson, Allan Wang

机构 * Robot Navigation and Perception Lab, AASS Research Center, Örebro University, Sweden（奥雷布罗大学机器人导航与感知实验室，AASS研究中心，瑞典）； Miraikan – The National Museum of Emerging Science and Innovation, Japan（日本新兴科学与创新国家博物馆——Miraikan）

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

AI总结提出HiCrowd分层框架，结合强化学习与模型预测控制，通过跟随人群流解决机器人冻结问题，在真实和合成数据集上提升导航效率与安全性。

Comments 2026 IEEE International Conference on Robotics and Automation (ICRA)

2509.16136 2026-06-09 cs.RO 版本更新 61%

Reward Evolution with Graph-of-Thoughts: A Bi-Level Language Model Framework for Reinforcement Learning

基于思维图的奖励进化：一种用于强化学习的双层语言模型框架

Changwei Yao, Xinzi Liu, Chen Li, Marios Savvides

机构 * Carnegie Mellon University（卡内基梅隆大学）； University of Tokyo（东京大学）

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

AI总结本文提出RE-GoT框架，结合LLM与VLM的图思维推理，通过任务分解和视觉反馈迭代优化奖励函数，实验表明在RoboGen和ManiSkill2任务中均优于现有方法。

Journal ref IEEE International Conference on Robotics and Automation (ICRA 2026)

2507.22345 2026-05-25 cs.RO 61%

A Reconfigured Wheel-Legged Robot for Enhanced Steering and Adaptability

一种增强转向能力和适应性的重构轮腿机器人

Zhicheng Song, Jinglan Xu, Chunxin Zheng, Yulin Li, Zhihai Bi, Jun Ma

机构 * Robotics and Autonomous Systems Thrust, The Hong Kong University of Science and Technology (Guangzhou)（机器人与自主系统方向，香港科技大学（广州））

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

AI总结提出一种名为FLORES的新型轮腿机器人，通过将前腿的髋关节横滚自由度替换为偏航自由度，并设计定制强化学习控制器，实现了高效转向和多地形适应。

Journal ref IEEE Robotics and Automation Letters, vol. 11, no. 6, pp. 7444-7451, June 2026

2512.08230 2026-04-16 cs.AI 61%

Empowerment Gain and Causal Model Construction: Children and adults are sensitive to controllability and variability in their causal interventions

赋能增益与因果模型构建：儿童和成人对干预可控性和变异性敏感

Eunice Yiu, Kelsey Allen, Shiry Ginosar, Alison Gopnik

机构 * Department of Psychology, University of California, Berkeley（加州大学伯克利分校心理学系）； Department of Computer Science, University of British Columbia（不列颠哥伦比亚大学计算机科学系）； Toyota Technological Institute at Chicago（芝加哥丰田技术研究所）

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

AI总结研究探讨了赋能增益在因果学习中的作用，通过实验验证儿童和成人如何利用赋能信号推断因果关系并设计干预措施。

Comments Accepted to Philosophical Transactions A, Special issue: World models, AGI, and the hard problems of life-mind continuity. Expected publication in 2026

2604.06943 2026-04-09 cs.RO 61%

Sustainable Transfer Learning for Adaptive Robot Skills

可持续的迁移学习用于适应性机器人技能

Khalil Abuibaid, Vinit Hegiste, Nigora Gafur, Achim Wagner, Martin Ruskowski

机构 * Chair of Machine Tools and Control System, RPTU University Kaiserslautern-Landau（机床与控制系统教席，莱茵兰-普法尔茨凯泽斯劳滕-兰道大学）； Innovative Factory Systems, German Institute of Artificial Intelligence（创新工厂系统，德国人工智能研究所）

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

AI总结本文研究了不同机器人平台间的策略迁移，通过强化学习完成peg-in-hole任务，探讨零样本迁移、微调和从头训练的效果，发现微调能显著提升性能，降低训练时间，支持可持续的机器人学习。

Comments Published in RAAD 2025 (Springer). 7 pages, 5 figures

Journal ref Advances in Service and Industrial Robotics, RAAD 2025, Springer, 2025, pp. 389-397

2011.01882 2026-04-07 cs.RO cs.GT 61%

Secure Planning Against Stealthy Attacks via Model-Free Reinforcement Learning

通过模型无关强化学习实现对隐蔽攻击的安全规划

Alper Kamil Bozkurt, Yu Wang, Miroslav Pajic

机构 * Duke University（杜克大学）

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

AI总结本文提出利用模型无关强化学习在未知随机环境中实现安全规划，通过将攻击者与控制器视为博弈双方，以线性时序逻辑公式表达其目标，解决在未知环境中满足LTL公式的问题。

Journal ref 2021 IEEE International Conference on Robotics and Automation (ICRA), Xi'an, China, 2021, pp. 10656-10662

2603.23182 2026-03-25 cs.RO cs.SY eess.SY 61%

Path Planning and Reinforcement Learning-Driven Control of On-Orbit Free-Flying Multi-Arm Robots

轨道自由飞行多臂机器人的路径规划与强化学习驱动控制

Álvaro Belmonte-Baeza, José Luis Ramón, Leonard Felicetti, Miguel Cazorla, Jorge Pomares

机构 * University of Alicante（阿利坎特大学）； Cranfield University（克兰菲尔德大学）

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

AI总结本文提出一种结合轨迹优化与强化学习的混合方法，用于自由飞行多臂机器人在在轨服务场景中的路径规划与控制。通过实验验证，该方法在表面运动和自由浮动场景中均优于传统策略，提升了运动平滑度、安全性和效率。

Comments Accepted for publication in The International Journal of Robotics Research (23-Mar-2026)

2603.07800 2026-03-10 cs.RO 61%

Preference-Conditioned Reinforcement Learning for Space-Time Efficient Online 3D Bin Packing

基于偏好条件的强化学习用于空间时间高效的在线3D装箱

Nikita Sarawgi, Omey M. Manyar, Fan Wang, Thinh H. Nguyen, Daniel Seita, Satyandra K. Gupta

机构 * Viterbi School of Engineering, University of Southern California（美国南加州大学维特比工程学院）； Amazon Robotics（亚马逊机器人）

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

AI总结 STEP方法通过偏好条件强化学习，在保持装箱密度的同时将操作时间减少44%。

Comments 8 pages, 5 figures. Accepted to IEEE International Conference on Robotics and Automation 2026. Project Website: https://step-packing.github.io

1. 模仿学习与强化学习 4115 篇

Autonomous Quadrotor Landing using Deep Reinforcement Learning

OptLayer - Practical Constrained Optimization for Deep Reinforcement Learning in the Real World

Modular Multi-Objective Deep Reinforcement Learning with Decision Values

Reset-free Trial-and-Error Learning for Robot Damage Recovery

RAIL: Risk-Averse Imitation Learning

Leave no Trace: Learning to Reset for Safe and Autonomous Reinforcement Learning

Learning Unmanned Aerial Vehicle Control for Autonomous Target Following

An Optimal Online Method of Selecting Source Policies for Reinforcement Learning

DropoutDAgger: A Bayesian Approach to Safe Imitation Learning

Learning Control for Air Hockey Striking using Deep Reinforcement Learning

Actor-Critic Reinforcement Learning with Simultaneous Human Control and Feedback

Reinforcement Learning for Pivoting Task

Reinforcement Learning With Temporal Logic Rewards

Vulnerability of Deep Reinforcement Learning to Policy Induction Attacks

RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning

Adaptive Probabilistic Trajectory Optimization via Efficient Approximate Inference

Reinforcement Learning for Semantic Segmentation in Indoor Scenes

Learning Deep Control Policies for Autonomous Aerial Vehicles with MPC-Guided Policy Search

A Real-Time Model-Based Reinforcement Learning Architecture for Robot Control

Publishing Identifiable Experiment Code And Configuration Is Important, Good and Easy

Reflex First, Reflect Later: Latency-Aware Embodied LLM Agents for Dynamic Response

Simulation of Surgical Suturing Using Position-Based Dynamics and the Material Point Method for Robot Reinforcement Learning

HiCrowd: Hierarchical Crowd Flow Alignment for Dense Human Environments

Reward Evolution with Graph-of-Thoughts: A Bi-Level Language Model Framework for Reinforcement Learning

A Reconfigured Wheel-Legged Robot for Enhanced Steering and Adaptability

Empowerment Gain and Causal Model Construction: Children and adults are sensitive to controllability and variability in their causal interventions

Sustainable Transfer Learning for Adaptive Robot Skills

Secure Planning Against Stealthy Attacks via Model-Free Reinforcement Learning

Path Planning and Reinforcement Learning-Driven Control of On-Orbit Free-Flying Multi-Arm Robots

Preference-Conditioned Reinforcement Learning for Space-Time Efficient Online 3D Bin Packing