机器人 / 具身智能 - arXivDaily 专题

1910.13399 2019-10-30 cs.RO cs.AI cs.LG 56%

Robust Model-free Reinforcement Learning with Multi-objective Bayesian Optimization

Matteo Turchetta, Andreas Krause, Sebastian Trimpe

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

Comments Submitted to IEEE Conference on Robotics and Automation 2020 (ICRA)

1910.07615 2019-10-18 cs.RO cs.AI cs.CV 56%

Conditional Driving from Natural Language Instructions

Junha Roh, Chris Paxton, Andrzej Pronobis, Ali Farhadi, Dieter Fox

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

Comments Accepted by the 3rd Conference on Robot Learning, Osaka, Japan (CoRL 2019)

1910.04365 2019-10-11 cs.RO cs.AI cs.LG 56%

Asking Easy Questions: A User-Friendly Approach to Active Reward Learning

Erdem Bıyık, Malayandi Palan, Nicholas C. Landolfi, Dylan P. Losey, Dorsa Sadigh

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

Comments Proceedings of the 3rd Conference on Robot Learning (CoRL), October 2019

1910.02646 2019-10-09 cs.RO cs.AI cs.LG cs.SY eess.SY 56%

Riemannian Motion Policy Fusion through Learnable Lyapunov Function Reshaping

Mustafa Mukadam, Ching-An Cheng, Dieter Fox, Byron Boots, Nathan Ratliff

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

Comments Conference on Robot Learning (CoRL), 2019

1909.03772 2019-09-12 cs.LG cs.AI cs.RO stat.ML 56%

A Survey on Reproducibility by Evaluating Deep Reinforcement Learning Algorithms on Real-World Robots

Nicolai A. Lynnerup, Laura Nolling, Rasmus Hasle, John Hallam

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

Comments Appears in Proceedings of the Third Conference on Robot Learning (CoRL 2019). Companion source code at https://github.com/dti-research/SenseActExperiments/

1908.02999 2019-08-09 cs.RO cs.CV cs.LG cs.MA 56%

Learning Vision-based Flight in Drone Swarms by Imitation

Fabian Schilling, Julien Lecoeur, Fabrizio Schiano, Dario Floreano

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

Comments 8 pages, 8 figures, accepted for publication in the IEEE Robotics and Automation Letters (RA-L) on July 28, 2019. arXiv admin note: substantial text overlap with arXiv:1809.00543

1806.02501 2018-06-08 cs.RO cs.AI cs.LG 56%

Simplifying Reward Design through Divide-and-Conquer

Ellis Ratner, Dylan Hadfield-Menell, Anca D. Dragan

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

Comments Robotics: Science and Systems (RSS) 2018

1711.10055 2018-03-23 cs.AI cs.LG cs.RO 56%

Risk-sensitive Inverse Reinforcement Learning via Semi- and Non-Parametric Methods

Sumeet Singh, Jonathan Lacotte, Anirudha Majumdar, Marco Pavone

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

Comments Submitted to International Journal of Robotics Research; Revision 1: (i) Clarified minor technical points; (ii) Revised proof for Theorem 3 to hold under weaker assumptions; (iii) Added additional figures and expanded discussions to improve readability

1710.05958 2017-10-18 cs.LG cs.AI cs.CV 56%

Gradient-free Policy Architecture Search and Adaptation

Sayna Ebrahimi, Anna Rohrbach, Trevor Darrell

专题命中模仿学习与强化学习：分类 cs.AI、cs.CV、cs.LG；robot learning(comments)

Comments Accepted in Conference on Robot Learning, 2017

1709.08430 2017-09-26 cs.RO cs.AI cs.LG 56%

Towards continuous control of flippers for a multi-terrain robot using deep reinforcement learning

Giuseppe Paolo, Lei Tai, Ming Liu

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

Comments 12 pages, single column, submitted to International Journal of Robotics and Automation (IJRA)

2503.14557 2025-11-18 cs.AI cs.MA cs.RO 54%

Generating Causal Explanations of Vehicular Agent Behavioural Interactions with Learnt Reward Profiles

Rhys Howard, Nick Hawes, Lars Kunze

机构 * Oxford Robotics Institute, Dept. of Eng. Sci., University of Oxford（牛津大学机器人研究所）

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

Comments 8 Pages, 5 Figures, To be published in the Proceedings of the 2025 IEEE International Conference on Robotics & Automation, Initial upload of accepted paper

Journal ref 2025 IEEE International Conference on Robotics and Automation (ICRA), Atlanta, GA, USA, 2025

2403.19946 2024-04-01 cs.RO cs.AI 54%

A Peg-in-hole Task Strategy for Holes in Concrete

André Yuji Yasutomi, Hiroki Mori, Tetsuya Ogata

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

Comments Published in 2021 IEEE International Conference on Robotics and Automation (ICRA) on 30 May 2021

Journal ref 2021 IEEE International Conference on Robotics and Automation (ICRA), Xi'an, China, 2021, pp. 2205-2211

2312.16438 2024-04-01 cs.RO cs.AI 54%

Visual Spatial Attention and Proprioceptive Data-Driven Reinforcement Learning for Robust Peg-in-Hole Task Under Variable Conditions

André Yuji Yasutomi, Hideyuki Ichiwara, Hiroshi Ito, Hiroki Mori, Tetsuya Ogata

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

Comments Published in IEEE Robotics and Automation Letters on 08 February 2023

Journal ref IEEE Robotics and Automation Letters, vol. 8, issue 3, pp. 1834-1841, 2023

2307.05832 2023-11-21 cs.CV cs.AI 54%

Bag of Views: An Appearance-based Approach to Next-Best-View Planning for 3D Reconstruction

Sara Hatami Gazani, Matthew Tucsok, Iraj Mantegh, Homayoun Najjaran

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

Comments Published in IEEE Robotics and Automation Letters (RA-L)

Journal ref IEEE Robotics and Automation Letters (RA-L), 2023

2302.07337 2023-08-28 cs.RO cs.AI cs.GT cs.MA 54%

Graph Attention Multi-Agent Fleet Autonomy for Advanced Air Mobility

Malintha Fernando, Ransalu Senanayake, Heeyoul Choi, Martin Swany

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

Comments Accepted to Robotics: Science and Systems, 2023. 14 pages, 13 figures, 3 tables

Journal ref Robotics: Science and Systems, 2023

2209.13220 2023-07-18 cs.RO cs.AI cs.LO 54%

Exploiting Transformer in Sparse Reward Reinforcement Learning for Interpretable Temporal Logic Motion Planning

Hao Zhang, Hao Wang, Zhen Kan

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

Comments IEEE Robotics and Automation Letters

Journal ref IEEE Robotics and Automation Letters, 2023

2206.10101 2022-09-01 cs.LG cs.AI 54%

Model-Based Imitation Learning Using Entropy Regularization of Model and Policy

Eiji Uchibe

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

Comments This is a preprint version of the paper to appear at IEEE Robotics and Automation Letters (RA-L). The final journal version is downloadable from https://doi.org/10.1109/LRA.2022.3196139

Journal ref IEEE Robotics and Automation Letters, Volume 7, Issue 4, pages 10922-10929, October, 2022

2008.00524 2022-03-09 cs.RO cs.LG 54%

Interactive Imitation Learning in State-Space

Snehal Jauhri, Carlos Celemin, Jens Kober

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

Comments Presented at the 4th Conference on Robot Learning (CoRL) 2020, 11 pages, 4 figures

Journal ref Proceedings of the 2020 Conference on Robot Learning, PMLR 155:682-692

2608.25402 2026-08-27 cs.NI 新提交 50%

Lightweight AI for UAV-Mounted RIS: An Overview

用于无人机搭载可重构智能表面（RIS）的轻量级AI：综述

Sherief Hashima, Kohei Hatano, Eiji Takimoto, Mohamed Rihan, Basem. M. Elhalawany, Hamada Rizk

专题命中模仿学习与强化学习：manipulation(abstract)

AI总结本文综述用于无人机搭载RIS的轻量级AI技术，涵盖RL、FL等方法，分析现有工作与权衡，确定研究挑战并通过案例展示MAB方案对系统性能的影响。

2603.28533 2026-08-11 cs.CL 版本更新 50%

GraphWalker: Agentic Knowledge Graph Question Answering via Synthetic Trajectory Curriculum

GraphWalker: 通过合成轨迹课程实现基于知识图谱的代理问答

Shuwen Xu, Yao Xu, Jiaxiang Liu, Chenhao Yuan, Wenshuo Peng, Jun Zhao, Kang Liu

机构 * School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）； The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所复杂系统认知与决策智能重点实验室）； Department of Electronic Engineering, Tsinghua University（清华大学电子工程系）

专题命中模仿学习与强化学习：navigation(abstract)

AI总结 GraphWalker通过自动化轨迹合成和分阶段微调解决知识图谱问答中的探索与泛化问题，提升轻量强化学习性能，在CWQ和WebQSP上取得最优效果。

Comments COLM 2026

2605.04902 2026-08-06 cs.DB 版本更新 50%

AegisTS: A Hierarchical Agentic AI System with Reinforcement Learning for Multivariate Time Series Data Cleaning

AegisTS: 一种基于强化学习的层次化智能体系统用于多变量时间序列数据清洗

Yuhan Shi, Yuanyuan Yao, Lu Chen, Mourad Khayati, Yushuai Li, Tianyi Li

专题命中模仿学习与强化学习：navigation(abstract)

AI总结针对多变量时间序列中同时存在的多种质量问题，提出基于强化学习的层次化智能体系统AegisTS，通过联合优化问题处理顺序和清洗模型选择，实现无需真值的高效数据清洗。

2607.21145 2026-08-05 eess.SY cs.SY 版本更新 50%

Deep Reinforcement Learning for Adaptive Gain Tuning in Control of Teleoperation Manipulators with Joint Flexibility and Time-Varying Delays

具有关节灵活性和时变延迟的遥操作机器人控制中自适应增益调整的深度强化学习

Armin Attarzadeh, Mohammad Ali Ghaemifar, Alireza Khanzadeh, Soheil Ganjefar

专题命中模仿学习与强化学习：robotic(abstract)

AI总结针对含有关节灵活性和时变延迟的遥操作机器人控制问题，提出结合稳定P+d控制器与基于TD3算法的深度强化学习智能体的混合控制方法，可实时调整增益减少振动，为相关遥操作系统提供实用方案。

Comments 7 pages, 6 figures. Source code available at: https://github.com/ArminAttarzadeh/DRL-Controller-Gain-Tuner

2607.26697 2026-07-30 physics.acc-ph 新提交 50%

Reinforcement Learning applied to Optimization of LHC beams in the CERN Proton Synchrotron

强化学习在欧洲核子中心质子同步加速器（CERN PS）LHC束流优化中的应用

Joel Axel Wulff, Alexandre Lasheen

专题命中模仿学习与强化学习：manipulation(abstract)

AI总结本研究将卷积神经网络与Soft-Actor-Critic强化学习智能体结合，实现CERN PS纵向三重分裂的自动化优化，其2025年部署的自主控制器为CERN注入器复合体首批强化学习束流质量优化系统之一。

Comments 15 pages, 14 figures, pre-print

2607.22541 2026-07-28 math.OC cs.NA math.NA math.PR 新提交 50%

SDE Guided Monte Carlo Reinforcement Learning: A Stochastic Maximum Principle Approach for Robust Decision Making in Noisy Environments

随机微分方程引导的蒙特卡罗强化学习：一种用于噪声环境中稳健决策的随机最大值原理方法

Juncai Wang

专题命中模仿学习与强化学习：navigation(abstract)

AI总结研究探讨SMP定性最优条件能否指导稳健表格型强化学习算法，提出SDE-MC-AC框架，通过建立对应关系及自适应温度调度整合多种方法，经实验验证了相关假设，揭示导航模式，为随机最优控制与强化学习搭建桥梁。

2607.18481 2026-07-22 cs.CL cs.IR 新提交 50%

Search-on-Graph-R1: Training Large Language Models to Search Knowledge Graphs with Reinforcement Learning

搜索图-R1：使用强化学习训练大型语言模型以搜索知识图谱

Jia Ao Sun, Hao Yu, Fengran Mo, Zhan Su, Yuchen Hui, Bang Liu, Jian-Yun Nie

机构 * Université de Montréal（蒙特利尔大学）； Mila – Québec AI Institute（米拉-魁北克人工智能研究所）； McGill University（麦吉尔大学）； Halmstad University College（哈尔姆斯塔德大学学院）

专题命中模仿学习与强化学习：navigation(abstract)

AI总结研究针对知识图谱问答部署成本高的问题，提出搜索图-R1，通过监督微调与强化学习，将导航内化到8B模型。核心是用黄金SPARQL查询搭建教师遍历答案路径。该模型在多个数据集上超越前沿语言模型，推理无需辅助模块，训练无需语言模型评判，且训练阶段互补，可跨模型家族迁移。

2607.17635 2026-07-21 eess.SY cs.SY 新提交 50%

On Optimal Event-Triggered Distributed Control for Stochastic Multi-Agent Systems via Reinforcement Learning

基于强化学习的随机多智能体系统最优事件触发分布式控制

Ziming Wang, Bingbing Li, Karl H. Johansson, Apostolos I. Rikos

专题命中模仿学习与强化学习：robotic(abstract)

AI总结针对含随机不确定性的多智能体系统，提出基于强化学习的最优分布式控制算法，采用 actor-critic-identifier 结构，用低通滤波器和混合事件触发控制策略，经稳定性证明，在仿真中验证正确性并与非最优算法比较突出优势。

2607.10830 2026-07-14 cs.CR 新提交 50%

Automated Stealthy Wear-Out Attack on Digital Twins With Deep Reinforcement Learning

基于深度强化学习的数字孪生自动隐身磨损攻击

Joshua Haworth, Aryan Pasikhani, George Pavlides, Prosanta Gope, John Clark

专题命中模仿学习与强化学习：robotic(abstract)

AI总结研究利用深度强化学习对数字孪生进行隐身磨损攻击，通过操纵控制信号加速特定关节磨损并躲避检测。测试多种算法发现SAC性能最佳，在工业环境中用UR10e机器人手臂评估，证明攻击有效，强调了该攻击对数字孪生环境的风险及防御需求。

2607.02533 2026-07-07 eess.SP cs.SY eess.SY 新提交 50%

Centralized PPO-Based DRL for Multi-UAV-BS Positioning and Trajectory Optimization in Disaster Response Networks

基于集中式近端策略优化的深度强化学习用于灾害响应网络中的多无人机基站定位与轨迹优化

Azim Akhtarshenas, Mario Rico Ibanez, Matteo Bernabe, David Lopez-Perez, Merouane Debbah

专题命中模仿学习与强化学习：navigation(abstract)

AI总结针对灾害场景中多无人机基站定位与轨迹优化问题，扩展集中学习框架，将其建模为马尔可夫决策过程，用近端策略优化的深度强化学习求解，能协调多无人机基站并适应异构用户移动模式。

Comments Submitted to IEEE Transactions on Machine Learning in Communications and Networking (TMLCN). Dataset available at https://doi.org/10.5281/zenodo.20720697 and code available at https://github.com/aakhtarshenas/centralized-ppo-multi-uav-bs

2607.02171 2026-07-03 cond-mat.stat-mech cond-mat.soft 新提交 50%

Theory of collective learning in populations of adaptive agents

自适应智能体群体中的集体学习理论

Gerhard Jung, Johann Asnacios, Misaki Ozawa, Olivier Dauchot, Eric Bertin

专题命中模仿学习与强化学习：robotics(abstract)

AI总结通过扩展动力学理论，研究同质活性智能体群体通过交换策略和记忆进行分散式学习以实现宏观目标的过程，推导出策略分布的演化方程，并揭示有效奖励函数的关键作用。

2606.31372 2026-07-01 cs.SE 新提交 50%

Failure-Based Testing for Deep Reinforcement Learning Agents

基于失败的深度强化学习智能体测试方法

Weibin Lin, Jiangtao Meng, Zheng Zheng

专题命中模仿学习与强化学习：robotic(abstract)

AI总结针对深度强化学习智能体测试中奖励信号失效的问题，提出一种基于任务失败洞察的黑盒测试方法PRT，通过优先测试高难度任务来提升故障检测效率，在四个基准上测试成本降低超50%。

Comments 22 pages

1. 模仿学习与强化学习 4134 篇

Robust Model-free Reinforcement Learning with Multi-objective Bayesian Optimization

Conditional Driving from Natural Language Instructions

Asking Easy Questions: A User-Friendly Approach to Active Reward Learning

Riemannian Motion Policy Fusion through Learnable Lyapunov Function Reshaping

A Survey on Reproducibility by Evaluating Deep Reinforcement Learning Algorithms on Real-World Robots

Learning Vision-based Flight in Drone Swarms by Imitation

Simplifying Reward Design through Divide-and-Conquer

Risk-sensitive Inverse Reinforcement Learning via Semi- and Non-Parametric Methods

Gradient-free Policy Architecture Search and Adaptation

Towards continuous control of flippers for a multi-terrain robot using deep reinforcement learning

Generating Causal Explanations of Vehicular Agent Behavioural Interactions with Learnt Reward Profiles

A Peg-in-hole Task Strategy for Holes in Concrete

Visual Spatial Attention and Proprioceptive Data-Driven Reinforcement Learning for Robust Peg-in-Hole Task Under Variable Conditions

Bag of Views: An Appearance-based Approach to Next-Best-View Planning for 3D Reconstruction

Graph Attention Multi-Agent Fleet Autonomy for Advanced Air Mobility

Exploiting Transformer in Sparse Reward Reinforcement Learning for Interpretable Temporal Logic Motion Planning

Model-Based Imitation Learning Using Entropy Regularization of Model and Policy

Interactive Imitation Learning in State-Space

Lightweight AI for UAV-Mounted RIS: An Overview

GraphWalker: Agentic Knowledge Graph Question Answering via Synthetic Trajectory Curriculum

AegisTS: A Hierarchical Agentic AI System with Reinforcement Learning for Multivariate Time Series Data Cleaning

Deep Reinforcement Learning for Adaptive Gain Tuning in Control of Teleoperation Manipulators with Joint Flexibility and Time-Varying Delays

Reinforcement Learning applied to Optimization of LHC beams in the CERN Proton Synchrotron

SDE Guided Monte Carlo Reinforcement Learning: A Stochastic Maximum Principle Approach for Robust Decision Making in Noisy Environments

Search-on-Graph-R1: Training Large Language Models to Search Knowledge Graphs with Reinforcement Learning

On Optimal Event-Triggered Distributed Control for Stochastic Multi-Agent Systems via Reinforcement Learning

Automated Stealthy Wear-Out Attack on Digital Twins With Deep Reinforcement Learning

Centralized PPO-Based DRL for Multi-UAV-BS Positioning and Trajectory Optimization in Disaster Response Networks

Theory of collective learning in populations of adaptive agents

Failure-Based Testing for Deep Reinforcement Learning Agents