arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

International Conference on Robotics and Automation · 会议 · Robotics

共收录 4585
2605.30617 2026-06-01 cs.RO math.OC

Exploiting Chordal Sparsity for Globally Optimal Estimation with Factor Graphs

利用弦稀疏性实现因子图的全局最优估计

Avinash Subramanian, Connor Holmes, Timothy D. Barfoot, Frank Dellaert, Frederike Dümbgen

机构 * College of Computing, Georgia Institute of Technology(佐治亚理工学院计算机学院) Robotics Institute, University of Toronto(多伦多大学机器人研究所) Department of Mechanical Engineering, Carnegie Mellon University(卡内基梅隆大学机械工程系)

AI总结 本文提出在GTSAM框架中自动构建凸半定规划松弛,并利用贝叶斯树分解加速求解,实现因子图的全局最优估计。

Journal ref ICRA 2026 WORKSHOP ON FRONTIERS OF OPTIMIZATION FOR ROBOTICS

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.30583 2026-06-01 cs.RO cs.PF

Caspar: CUDA Accelerator for Symbolic Programming with Adaptive Reordering

Caspar: 基于自适应重排序的符号编程CUDA加速器

Emil Martens, Aaron Miller, Matias Varnum, Annette Stahl

机构 * Norwegian University of Science and Technology(挪威科学与技术大学) Skydio

AI总结 提出Caspar库,通过自动生成优化CUDA内核,实现从Python符号表达式到GPU高性能运行时的桥梁,并在大规模BA数据集上实现5-20倍加速。

Comments Accepted at ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21013 2026-06-01 cs.RO

Notes-to-Self: Scratchpad Augmented VLAs for Memory Dependent Manipulation Tasks

笔记到自我:带草稿本的增强型VLA用于依赖记忆的操作任务

Sanjay Haresh, Daniel Dijkman, Apratim Bhattacharyya, Roland Memisevic

机构 * Qualcomm AI Research(高通AI研究)

AI总结 本文通过在视觉-语言-动作模型中加入语言草稿本来赋予其空间和时间记忆,从而提升其在依赖记忆的长时域操作任务上的泛化能力。

Comments To appear at ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.20795 2026-06-01 cs.RO

Learning Generalizable Robot Policy with Human Demonstration Video as a Prompt

以人类演示视频为提示学习可泛化的机器人策略

Xiang Zhu, Yichen Liu, Hezhong Li, Jianyu Chen

机构 * Tsinghua University, China(清华大学,中国) Shanghai Qi Zhi Institute, China(上海启智研究院,中国)

AI总结 提出两阶段框架,利用人类演示视频学习可泛化机器人策略,无需遥操作数据或微调即可执行新任务。

Comments Accepted to the IEEE International Conference on Robotics and Automation (ICRA), 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.29773 2026-05-29 cs.CV cs.AI cs.RO

Energy-Aware NECO for Single-Pass Pixel-wise Out-of-Distribution Detection in Semantic Segmentation

能量感知NECO:用于语义分割中单次逐像素分布外检测

Boyuan Zhang, Huanshan Huang, Yifei Cao

机构 * Ecole Polytechnique, Institut Polytechnique de Paris(巴黎理工学院高研院) CIAD, UTBM, Université Marie et Louis Pasteur(CIAD、UTBM、马吕斯·路易·巴斯蒂埃大学) U2IS, ENSTA, Institut Polytechnique de Paris(U2IS、ENSTA、巴黎理工学院)

AI总结 提出一种结合NECO几何比率和能量分数的混合方法,实现单次前向传播的逐像素分布外检测,在miniMUAD数据集上AUROC达0.8539,优于单独使用NECO或能量分数。

Comments 7 pages, 6 figures. Accepted at the ICRA 2026 Workshop on Long-term Deployments in the Wild (LoWi 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.17798 2026-05-29 cs.RO

SM2ITH: Safe Mobile Manipulation with Interactive Human Prediction via Task-Hierarchical Bilevel Model Predictive Control

SM2ITH:通过任务分层双层模型预测控制实现安全移动操作与人机交互预测

Francesco D'Orazio, Sepehr Samavi, Xintong Du, Siqi Zhou, Giuseppe Oriolo, Angela P. Schoellig

机构 * Department of Computer, Control and Management Engineering, of Sapienza University of Rome(意大利萨皮恩扎大学计算机、控制与管理工程系) University of Toronto Institute for Aerospace Studies (UTIAS) and the Vector Institute for Artificial Intelligence(多伦多大学航空航天研究所(UTIAS)和向量人工智能研究所) Learning Systems and Robotics lab at the Technical University of Munich and the Munich Institute for Robotics and Machine Intelligence (MIRMI)(慕尼黑技术大学学习系统与机器人实验室及慕尼黑机器人与机器智能研究所(MIRMI)) School of Computing Science, Faculty of Applied Sciences, Simon Fraser University(西蒙·弗雷泽大学应用科学学院计算机科学系)

AI总结 提出SM$^2$ITH框架,结合分层任务模型预测控制与双层优化的人机交互预测,实现动态人机环境中的安全高效移动操作。

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

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.04758 2026-05-29 cs.RO cs.AI cs.MA

ScheduleStream: Temporal Planning with Samplers for GPU-Accelerated Multi-Arm Task and Motion Planning & Scheduling

ScheduleStream: 基于采样器的时序规划用于GPU加速的多臂任务与运动规划及调度

Caelan Garrett, Fabio Ramos

机构 * NVIDIA Research Seattle Robotics Lab (SRL)(NVIDIA西雅图机器人实验室) University of Sydney(悉尼大学)

AI总结 提出ScheduleStream,首个通用框架,通过混合持续动作和领域无关算法,结合GPU加速采样器,实现多臂并行任务与运动规划及调度。

Comments Project website: https://schedulestream.github.io

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

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.29298 2026-05-29 cs.RO

MonoDuo: Using One Robot Arm to Learn Bimanual Policies

MonoDuo: 使用单机械臂学习双臂策略

Sandeep Bajamahal, Lawrence Yunliang Chen, Toru Lin, Zehan Ma, Jitendra Malik, Ken Goldberg

机构 * University of California, Berkeley(加州大学伯克利分校)

AI总结 提出MonoDuo框架,利用单臂机器人演示和人类协作数据,通过数据增强生成合成演示,训练双臂机器人策略,在五项任务中实现零样本部署和少样本微调,成功率高达70%。

Comments Accepted to appear in the 2026 IEEE International Conference on Robotics and Automation (ICRA), Vienna, Austria, 1-5 June 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.08142 2026-05-29 cs.RO

Multifingered force-aware control for humanoid robots

人形机器人的多指力感知控制

Pasquale Marra, Gabriele M. Caddeo, Ugo Pattacini, Lorenzo Natale

机构 * Humanoid Sensing and Perception, Istituto Italiano di Tecnologia, Genoa, Italy(人机感知与感知,意大利技术研究院,热那亚,意大利) DIBRIS, Università di Genova, Via All’Opera Pia, 13, Genoa, Italy(DIBRIS,热那亚大学,Via All’Opera Pia, 13,热那亚,意大利) MESH Facility, Istituto Italiano di Tecnologia, Genoa, Italy(MESH设施,意大利技术研究院,热那亚,意大利)

AI总结 提出一种基于力估计的多指手人形机器人控制方案,通过调整躯干、手臂、手腕和手指的运动重新分配力,以维持与物体的稳定接触,在平衡任务中成功率达82.7%。

Comments This work has been accepted for publication in ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.09976 2026-05-29 cs.RO

Masquerade: Learning from In-the-wild Human Videos using Data-Editing

Masquerade: 利用数据编辑从真实世界人类视频中学习

Marion Lepert, Jiaying Fang, Jeannette Bohg

机构 * Stanford University(斯坦福大学)

AI总结 提出Masquerade方法,通过编辑真实世界第一人称人类视频(估计3D手部姿态、修复手臂、叠加渲染双臂机器人)弥合视觉具身差距,并利用编辑后的视频预训练视觉编码器、微调扩散策略头,在三个长时程双臂厨房任务中实现比基线高5-6倍的泛化性能。

Comments Project website at https://masquerade-robot.github.io/

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

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.28726 2026-05-28 cs.RO cs.LG

How VLAs Fail Differently: Black-Box Action Monitoring Reveals Architecture-Specific Failure Signatures

VLA如何以不同方式失败:黑盒动作监控揭示架构特定的失败特征

Krishnam Gupta

机构 * Independent Research(独立研究)

AI总结 本文通过黑盒动作监控发现,视觉-语言-动作(VLA)架构在电机指令层面以根本不同且可预测的方式失败,并证明架构匹配的监控器选择至关重要。

Comments Accepted at IEEE ICRA 2026 Workshop "From Data to Decisions: VLA Pipelines for Real Robots", Vienna, June 2026. Non-archival workshop. 5 pages, 2 figures, 22 references

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.28468 2026-05-28 cs.RO

EIT-Pneumatic Hybrid Robotic Skin for Practical and Accurate Force Map Reconstruction

EIT-气动混合机器人皮肤用于实用且精确的力图重建

Junhwi Cho, Sunggyu Bae, Junghyeon Ma, Hyosang Lee, Jung Kim, Kyungseo Park

机构 * Mechanical Engineering Department, KAIST(韩国科学技术院机械工程系) Department of Robotics and Mechatronics Engineering, DGIST(大邱科学技术院机器人与机电工程系) Mechanical Engineering Department, TU/e(埃因霍温理工大学机械工程系)

AI总结 提出一种结合电阻抗断层成像(EIT)与气动触觉传感的混合机器人皮肤,通过Tikhonov正则化逆重建和逐垫气动校准,实现大面积精确触觉传感,并降低灵敏度不均匀性。

Comments 8 pages, 8 figures. Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026. J. Cho, S. Bae, J. Ma contributed equally

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.28412 2026-05-28 cs.RO cs.LG

Tactile-Proprioceptive Sensor Fusion for Contact Wrench Estimation in Whole-Body Physical Human-Robot Interaction

触觉-本体感觉传感器融合用于全身物理人机交互中的接触力估计

Junha Min, Junghyeon Ma, Jiwung Kwon, Sunggyu Bae, Joohyung Kim, Kyungseo Park

机构 * Department of Robotics and Mechatronics Engineering, DGIST (Daegu Gyeongbuk Institute of Science and Technology)(机器人与机电工程系,DGIST(大邱庆尚科学技术研究所)) Kinetic Intelligent Machine Lab (KIMLAB), University of Illinois Urbana-Champaign(动能智能机器实验室(KIMLAB),伊利诺伊大学厄巴纳-香槟分校)

AI总结 提出触觉-本体感觉融合框架,利用气动皮肤垫的触觉线索作为接触指示器,结合基于电机电流的本体感觉,通过时间卷积网络消除摩擦滞后,实现多轴接触力重建,提高物理人机交互的灵敏度和响应性。

Comments 8 pages, 6 figures. Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.28372 2026-05-28 cs.LG cs.RO

Teacher-Student Representational Alignment for Reinforcement Learning-Driven Imitation Learning

教师-学生表征对齐用于强化学习驱动的模仿学习

Meraj Mammadov, Pedro Zuidberg Dos Martires, Johannes Andreas Stork

机构 * Department of Computer Science(计算机科学系) Örebro University(奥雷布罗大学)

AI总结 提出一种通过自监督对比学习构建共享嵌入空间的方法,以减小教师和学生策略之间的不可模仿差距,从而提升学生策略性能。

Comments 6 pages, 5 figures. Accepted as an oral presentation at the RL4IL Workshop at ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.28352 2026-05-28 cs.RO

Magnet-Based Soft Robotic Skin Using a 3D-Printed Multi-Lattice Structure and CNN-Based Tactile Super-Resolution

基于磁体的软体机器人皮肤:使用3D打印多格点结构和CNN触觉超分辨率

Yunseong Bang, Joowon Park, Suan Sim, Youngjun Ryu, Sukho Park, Kyungseo Park

AI总结 提出一种集成多层软格点、霍尔效应传感器阵列和CNN触觉超分辨率模型的磁基机器人皮肤,通过格点参数调节实现机械柔顺性与传感特性的联合优化,并利用3D打印快速制造,实现接触位置和法向力的实时估计。

Comments 6 pages, 9 figures. Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026. Y. Bang and J. Park contributed equally

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.28312 2026-05-28 cs.RO cs.CV

EventShiftFlow: Towards Hardware-efficient FPGA-based Flow Estimation

EventShiftFlow:面向硬件高效的基于FPGA的流估计

Arianna Alonso Bizzi, Fernando Cladera, C. J. Taylor

机构 * University of Pennsylvania(宾夕法尼亚大学)

AI总结 提出一种基于事件相机的流估计方法,通过离散化事件、构建1位空间占用网格并并行评估速度假设,仅使用固定宽度整数逻辑实现,无需帧重建、浮点运算或迭代优化,适用于低延迟机器人感知。

Comments 10 pages, 5 figures. Accepted to the IEEE ICRA 2026 Workshop on Challenges and Opportunities of Neuromorphic Field Robotics and Automation

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.13003 2026-05-28 cs.RO cs.SY eess.SY

From Passive Monitoring to Active Defence: Resilient Control of Manipulators Under Cyberattacks

从被动监测到主动防御:网络攻击下机械臂的弹性控制

Gabriele Gualandi, Alessandro V. Papadopoulos

机构 * Department of Computer Science and Engineering, Mälardalen University(计算机科学与工程系,马尔默大学)

AI总结 针对虚假数据注入攻击(FDIA)下冗余机械臂的弹性控制问题,提出一种基于异常分数的主动控制级防御方法,通过单调函数衰减控制输入,显著降低攻击引起的末端执行器偏差,同时保证无攻击时的标称性能。

Comments v2: Accepted at ICRA 2026. Corrected minor typos, grammatical errors, and notation inconsistencies. Corrected the attacker's PD law in Sec. III-C: removed the feedforward acceleration term, viable only when the attacker assumes sufficient tracking precision; the active defence prevents this in our experiments, so only PD terms are used

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.03534 2026-05-28 cs.MA cs.LG cs.SY eess.SY stat.ML

Long-Term Mapping of the Douro River Plume with Multi-Agent Reinforcement Learning

基于多智能体强化学习的杜罗河羽流长期映射

Nicolò Dal Fabbro, Milad Mesbahi, Renato Mendes, João Borges de Sousa, George J. Pappas

机构 * University of Pennsylvania(宾夕法尼亚大学) Faculdade de Engenharia da Universidade do Porto(波尔图大学工程学院) Laboratório de Sistemas e Tecnologia Subaquática (LSTS)(水下系统与技术实验室) Laboratório Associado de Energia, Transportes e Aeronáutica (LAETA)(能源、运输与航空联合实验室)

AI总结 提出一种能量与通信高效的多智能体强化学习方法,结合时空高斯过程回归与多头Q网络控制器,实现多艘自主水下航行器对杜罗河羽流的长期(多天)映射,在Delft3D模拟中优于基准方法,且增加智能体数量可提升精度与续航。

Comments Accepted at the 2026 IEEE International Conference on Robotics and Automation

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.14075 2026-05-28 cs.RO cs.SY eess.SY

RCM Constraint-Consistent Dynamic Control in Surgical Robots

手术机器人中的RCM约束一致性动态控制

Yu Li, Hamid Sadeghian, Zewen Yang, Valentin Le Mesle, Sami Haddadin

机构 * Munich Institute of Robotics and Machine Intelligence, Technical University of Munich, Germany(慕尼黑机器人与机器智能研究所,慕尼黑技术大学,德国) Mohamed Bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE(穆罕默德·本·扎耶德人工智能大学,阿布扎比,阿联酋)

AI总结 将远程运动中心(RCM)建模为流变完整约束,并集成到基于投影的逆动力学控制器中,实现扭矩层面的约束一致控制,降低RCM残差并平滑扭矩曲线。

Comments Accepted at ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.13177 2026-05-28 cs.RO

ROOM: A Physics-Based Continuum Robot Simulator for Photorealistic Medical Datasets Generation

ROOM: 基于物理的连续体机器人模拟器,用于生成逼真的医学数据集

Salvatore Esposito, Matías Mattamala, Daniel Rebain, Francis Xiatian Zhang, Kevin Dhaliwal, Mohsen Khadem, Subramanian Ramamoorthy

机构 * University of Edinburgh, UK(爱丁堡大学,英国) University of British Columbia, Canada(不列颠哥伦比亚大学,加拿大)

AI总结 提出ROOM模拟框架,利用患者CT扫描生成多模态支气管镜训练数据,验证其在姿态估计和深度估计任务中的有效性。

Journal ref International Conference on Robotics and Automation 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.26828 2026-05-27 cs.RO

Learning Compositional Symbolic Task Rules from Demonstrations with Inductive Logic Programming

通过归纳逻辑编程从演示中学习组合符号任务规则

Oleh Borys, Karla Stepanova

机构 * Czech Institute of Informatics, Robotics and Cybernetics(捷克信息学、机器人学与自动控制研究所)

AI总结 提出一种基于归纳逻辑编程的分解学习方法,从演示中学习可解释、可重用且支持强泛化的符号任务规则。

Comments In: ICRA 2026 Workshop on Semantics for Reliable Robot Autonomy: From Environment Understanding and Reasoning to Safe Interaction, Vienna, 2026 In: ICRA 2026, International Joint Workshop on Ontologies, Semantic Maps and Autonomous Robotics Standardization (J-WOSMARS 2026), Vienna, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.24465 2026-05-27 cs.RO

Polymander II: an amphibious salamander-inspired robot with contact and flow sensors

Polymander II:一种带有接触和流量传感器的两栖蝾螈启发机器人

Qiyuan Fu, Sudong Lee, Andrea Grillo, Jonathan Arreguit, Louis Gevers, Josie Hughes, Auke J. Ijspeert

机构 * Biorobotics Laboratory, EPFL(生物机器人实验室,瑞士联邦理工学院) CREATE Lab, EPFL(CREATE实验室,瑞士联邦理工学院) Innobridge Services Sàrl(Innobridge Services公司)

AI总结 本文提出一种基于霍尔效应传感器的两栖机器人,用于感知足部接触力和侧向水动力,实现陆水环境感知与反馈控制。

Comments This work has been accepted for publication in the 2026 International Conference on Robotics and Automation (ICRA), Vienna, Austria

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.20255 2026-05-27 cs.LG cs.AI cs.HC cs.RO

Multi-Agent Reinforcement Learning for Safe Autonomous Driving Under Pedestrian Behavioral Uncertainty

行人行为不确定性下安全自动驾驶的多智能体强化学习

Prakash Aryan, Kaushik Raghupathruni, Timo Kehrer, Sebastiano Panichella

机构 * University of Bern(伯恩大学) AI4I, The Italian Institute of Artificial Intelligence(意大利人工智能研究所)

AI总结 本文使用多智能体近端策略优化(MAPPO)联合训练自动驾驶汽车和12个行人,通过隐藏的行人特质模拟乱穿马路行为,相比固定策略基线显著降低了碰撞率,并揭示了速度差异指标可用于检测未预期的乱穿马路行为。

Comments Accepted to ICRA 2026 Workshop "8th Workshop on Long-term Human Motion Prediction"

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.12918 2026-05-27 cs.CV

Radar-Camera BEV Multi-Task Learning with Cross-Task Attention Bridge for Joint 3D Detection and Segmentation

雷达-相机BEV多任务学习:用于联合3D检测与分割的跨任务注意力桥

Ahmet İnanç, Özgür Erkent

机构 * Hacettepe University(哈切特佩大学)

AI总结 提出CTAB(跨任务注意力桥)模块,通过共享BEV空间中的多尺度可变形注意力在检测和分割分支间交换特征,实现联合3D检测与分割的多任务学习,在nuScenes上提升分割性能且检测几乎不受影响。

Comments 8 pages, 5 figures, 3 Tables, Accepted at Radar in Robotics: New Frontiers workshop, at IEEE International Conference on Robotics & Automation (ICRA), 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.18384 2026-05-27 cs.RO cs.FL

LAD-VF: LLM-Automatic Differentiation Enables Fine-Tuning-Free Robot Planning from Formal Methods Feedback

LAD-VF:LLM自动微分实现基于形式化方法反馈的无微调机器人规划

Yunhao Yang, Junyuan Hong, Gabriel Jacob Perin, Zhiwen Fan, Li Yin, Zhangyang Wang, Ufuk Topcu

机构 * The University of Texas at Austin(德克萨斯大学奥斯汀分校) University of São Paulo(圣保罗大学) Texas A&M University(德克萨斯A&M大学) SylphAI

AI总结 提出LAD-VF框架,利用形式化验证反馈和LLM自动微分自动优化提示词,无需微调即可提升机器人规划任务中规范符合率,成功率从60%提升至90%以上。

Comments Presented at ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2009.11997 2026-05-27 cs.LG cs.AI cs.RO

Continual Model-Based Reinforcement Learning with Hypernetworks

基于超网络的连续模型强化学习

Yizhou Huang, Kevin Xie, Homanga Bharadhwaj, Florian Shkurti

机构 * Division of Engineering Science, University of Toronto, Canada(多伦多大学工程科学系) Department of Computer Science, University of Toronto, Canada(多伦多大学计算机科学系)

AI总结 提出HyperCRL方法,利用任务条件超网络在序列任务中持续学习动力学模型,避免重新训练并固定存储开销,在机器人 locomotion 和 manipulation 任务中优于现有持续学习方法。

Comments Updated link to project website in the abstract. 7 pages (+2 pages in appendix), 8 figures. In proceedings of the 2021 IEEE International Conference on Robotics and Automation

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.25942 2026-05-26 cs.CV cs.RO

LRDDv3: High-Resolution Long-Range Drone Detection Dataset with Range Information and Thermal Data

LRDDv3:具有距离信息和热数据的高分辨率远程无人机检测数据集

Knut Peterson, Zaid Mayers, Azmain Yousuf, Priontu Chowdhury, Asher Zaczepinski, Solmaz Arezoomandan, Reihaneh Maarefdoust, David Han

机构 * iMaPLe Research Lab, Drexel University(Drexel大学iMaPLe研究实验室) University of Maine(缅因大学)

AI总结 提出LRDDv3数据集,包含102,532张高分辨率远程RGB图像和29,630张配对IR图像,支持远程无人机检测,提供距离信息。

Comments 8 pages, 5 figures. Accepted to the 2026 IEEE International Conference on Robotics and Automation (ICRA)

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.25790 2026-05-26 cs.RO

HoLoArm: Deformable Arms for Collision-Tolerant Quadrotor Flight

HoLoArm: 用于碰撞容忍四旋翼飞行的可变形臂

Quang Ngoc Pham, Jonas Eschmann, Yang Zhou, Alejandro Ojeda Olarte, Giuseppe Loianno, Van Anh Ho

机构 * Japan Advanced Institute of Science and Technology(日本先进科学技术研究所) University of California Berkeley(加州大学伯克利分校) New York University(纽约大学)

AI总结 受蜻蜓翅膀结脉结构启发,提出具有柔性臂的四旋翼HoLoArm,结合强化学习控制策略实现被动变形与快速恢复,在高达7.6 m/s碰撞速度下保持稳定飞行。

Comments 8 pages, 15 figures, 1 table, Accepted at the IEEE Robotics and Automation Letters (RA-L) and the IEEE International Conference on Robotics and Automation (ICRA), 2026

Journal ref IEEE Robotics and Automation Letters, vol. 11, no. 3, pp. 3582-3589, March 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.25262 2026-05-26 cs.CV

Semantics-Guided Multimodal Masked Autoencoder Pretraining for 3D BEV Object Detection

语义引导的多模态掩码自编码器预训练用于3D BEV目标检测

Prabuddhi Wariyapperuma, Rajitha de Silva, Marc Hanheide, Thomas Bohné, Leonardo Guevara

机构 * University of Lincoln, Lincoln Centre for Autonomous Systems(林肯大学,林肯自主系统中心) University of Cambridge, Institute for Manufacturing, Department of Engineering(剑桥大学,制造研究所,工程系)

AI总结 提出语义引导的多模态掩码自编码器框架,通过语义引导的LiDAR体素掩码和辅助点语义解码分支,在预训练中注入语义信息,提升3D BEV目标检测性能。

Comments Accepted at the ICRA 2026 Workshop on Semantics for Reliable Robot Autonomy (SRRA) as a lightning talk and poster

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.06218 2026-05-26 cs.RO

Few-Shot Neural Differentiable Simulator: Real-to-Sim Rigid-Contact Modeling

少样本神经可微模拟器:真实到模拟的刚体接触建模

Zhenhao Huang, Siyuan Luo, Bingyang Zhou, Ziqiu Zeng, Jason Pho, Fan Shi

机构 * National University of Singapore(新加坡国立大学) Prana Lab(Prana实验室)

AI总结 提出一种结合解析公式物理一致性与图神经网络表示能力的少样本真实到模拟方法,通过少量真实数据校准解析模拟器生成大规模合成数据集,并引入基于网格的图神经网络隐式建模刚体前向动力学及碰撞检测的代理梯度,实现完全可微性,从而提升模拟保真度和策略学习效率。

Comments Accepted in ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏