MEKAWIRE
研究

论文

机器人与具身智能领域的最新论文，以及每篇对产业的意义。

73 篇论文
主题全部操作 76VLA 65导航 35机器人综合 32机器人学习 27感知 22人形机器人 21仿真 15运动控制 13安全 11World Action Models 4World-action model 3Aerial 2Aerial manipulation 2Dexterous manipulation 2Sim-to-Real Manipulation 2Social Navigation 2Vision-Language Navigation 23D Assembly 13DGS dataset 15G Simulation 1AMR Operations 1Active vision manipulation 1Adaptive MPC 1Adaptive action horizons 1Adaptive intervention 1Aerial Manipulation 1Aerial navigation 1Agricultural Robotics 1Articulation Estimation 1Autonomous Driving 1Autonomous Exploration 1Battery-aware RL 1Belief-space planning 1Bimanual Manipulation 1Byzantine localization 1Canal inspection 1Co-speech Humanoid 1Compositional Manipulation 1Construction navigation 1Contact-Rich Manipulation 1Contact-rich Manipulation 1Continual reinforcement learning on top of robot foundation models 1Continuum sensing 1Cross-embodiment transfer of humanoid whole-body controllers 1Curriculum RL 1Dexterous Manipulation 1Distributed Manipulation 1Distribution Shift 1Driving Simulation 1Efficient VIO 1Egocentric data 1Egocentric video 1Embodied Agents 1Embodied policy 1Empowerment MPC 1Endoscopic navigation 1Endovascular robotics 1Evaluating robot world models against real executions 1Event-camera VO 1Exoskeletons 1Failure recovery 1Federated 3D Perception 1Fisheye detection 1Floorplan SLAM 1Geometry-aware manipulation 1Grasping 1Grounding 1Humanoid Control 1Humanoid RL 1Humanoid Tool Use 1Humanoid loco-manipulation 1Humanoid safety 1Humanoid whole-body control 1Imitation Learning 1JEPA World Models 1Lab automation 1Laboratory Manipulation 1Latent World Models 1Legged Manipulation 1LiDAR Security 1Lifelong RL 1MAPF 1MAV guidance 1Manipulation LfD 1Manipulation Safety 1Marine simulation 1Memory-centric VLA 1Miniature Locomotion 1Mining Perception 1Mobile manipulation 1Model-based RL 1Monocular SSC 1Motion Planning 1Multi-Agent RL 1Multi-Robot Exploration 1Multi-Robot Manipulation 1Multi-UAV Exploration 1Multi-UAV State Estimation 1Multi-agent RL 1Multi-agent motion planning 1Multi-camera odometry 1Multi-fidelity RL 1Multi-robot manipulation 1Multi-skill Manipulation 1Multi-view perception 1Multimodal simulation 1Navigation & Mapping 1Object Pose Estimation 1Object Search 1Panoramic 3D Detection 1Panoramic segmentation 1Pedestrian tracking 1PersonNav 1Planning 1Policy Optimization 1Preference Learning 1Quadruped locomotion 1Real-time VLA 1Real-to-sim 1Reinforcement learning for industrial assembly 1Representation alignment 1Robot-Use Agents 1Runtime Verification 1Safe Generative Planning 1Safe RL-MPC 1Safe VLA 1Safety monitoring 1Safety-critical control 1Sim-to-real Navigation 1Sim-to-real manipulation 1Simulation geometry 1Skill Discovery 1Social navigation 1Soft Pneumatic Actuators 1Soft manipulation 1Sonar SLAM 1Streaming 3D reconstruction 1Submovement decomposition 1Surgical 1Surgical robotics 1Tactile Exploration 1Tactile VLA 1Temporal manipulation 1Terrain traversability 1Text-to-3D Policy 1Tool Co-Design 1Trajectory Correction 1Traversability 1UAV Navigation 1UAV Swarms 1UAV traffic monitoring 1Underwater Robotics 1Underwater active mapping 1Underwater perception 1Unified world-action model 1VLA Conditioning 1VLA Defense 1VLA Robustness 1VLA evaluation 1VLA manipulation 1VLA robustness 1Video world models 1Visual Generalization 1Visual Grounding 1Visual Reward Models 1Visuomotor Fusion 1Visuotactile manipulation 1World Models 1World models 1World-Action Models 1
难度不限入门进阶专业
暂无中文版，显示英文原文。
按技术筛选：VLA 视觉语言动作模型，清除
arXiv
VLA专业

Recovering Low-Bit Failures in Vision-Language-Action Policies via Quantization-Conditioned Reinforcement Learning

Seongmin Park, Jaewon Lee, Yujun Heo, Yoonsik Kim, Seheon Choi, Hyoung-rock Kim, Jungwook Choi

Quantization lowers deployment costs of vision-language-action (VLA) policies but can compromise task completion. Across five VLA models and 4,050 real-world trials, we systematically reveal that quantization-aware training (QAT) generally improves on post-training quantization (PTQ) but leaves substantial gaps to full precision (FP) on long-horizon tasks and under visual and semantic shifts, while its imitation-based objective does not explicitly use rollout success or partial completion.

VLARL
阅读论文全文（PDF）
arXiv
VLA专业

Vision-Predictor-Action: Towards Efficient Embodied AI via Neuro-Symbolic JEPA Predictors

Yeonseok Lee

Vision-Language-Action (VLA) models commonly generate robot actions token by token through Large Language Model (LLM) decoders, causing inference latency to grow with action sequence length and limiting closed-loop control frequency. We present Goal-Conditioned Vision-Predictor-Action (VPA), which replaces autoregressive decoders with a lightweight latent control loop.

VLA
阅读论文全文（PDF）
arXiv
VLA专业

RESCUE: Is This Alarm Worth Acting On? Counterfactual Evaluation of VLA Runtime Monitors

David Ahmedt-Aristizabal, Abdelwahed Khamis, Mohammad Ali Armin, Lars Petersson

Runtime failure monitors for robot policies are typically evaluated by how accurately they distinguish failed from successful executions. Yet an accurate alarm may arrive too late for recovery, while an unnecessary intervention may disrupt an execution that would otherwise succeed.

VLA
阅读论文全文（PDF）
arXiv
VLA专业

Reflect-VLA: Internalizing Reflective Guidance via Dual-Path On-Policy Self-Distillation for Autonomous Driving

Yuechen Luo, He Chong, Fang Li, Shaoqing Xu, Hexin Zhang, Ruiqi Zhang, Ziying Song, Lei Yang, Zhi-Xin Yang, Fuxi Wen

Reinforcement Learning (RL) substantially improves planning performance in autonomous-driving Vision-Language-Action (VLA) models. However, existing RL methods rely on scalar trajectory rewards that measure overall quality but fail to identify specific planning errors or provide actionable correction signals.

VLARL
阅读论文全文（PDF）
arXiv
VLA专业

Forward Dynamics Models

Zipei Ma, Junzhe Jiang, Jiahui Zhang, Juntong Zhang, Chun Gu, Bozhou Zhang, Jiankang Deng, Xiatian Zhu, Li Zhang

Vision-language-action (VLA) models and world action models (WAMs) have advanced robotic policy learning by building on pretrained vision-language models (VLMs) and video generative models (VGMs), respectively. Yet VLAs learn without explicit supervision from future world states, while WAMs that condition actions on predicted futures incur extra inference cost and let visual prediction errors propagate into actions.

VLA
阅读论文全文（PDF）
arXiv
VLA专业

RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation

Xuhui Kang, Sung-Wook Lee, Yan Zhuang, Tianle Zhong, Yen-Ling Kuo

Pre-trained robot policies, such as Vision-Language-Action (VLA) models, provide a strong foundation for generalist robotic manipulation, but adapting them to new tasks with only a few demonstrations remains challenging. Directly fine-tuning a pre-trained policy often achieves strong in-distribution performance, yet it can distort the representations learned during pre-training and weaken the model's generalization ability under out-of-distribution shifts.

VLA
Foundation
阅读论文全文（PDF）
arXiv
VLA专业

Noise Space Learning for VLA Adaptation

Junjie Lu, Chuheng Zhang, Yucen Wang, Yuhua Jiang, Xinyao Qin, Min Xu, Li Zhao

Pretrained vision-language-action (VLA) models encode broad manipulation priors, yet adapting them to specific tasks requires balancing limited robot interaction, computational resources, and safety constraints. Noise space learning (NSL) adapts flow-matching-based VLAs by learning a context-conditioned distribution over noise while keeping the pretrained VLA fixed.

VLA
阅读论文全文（PDF）
arXiv
VLA专业

Beyond Actions: Learning Future Operation Targets for Vision-Language-Action Models

Min Lin, Bokai Zhou, Jianqi Lin, Kaiyuan Yang, Yicheng Chang, Junjie Yu, Xiaofan Liu, Jianhua Han, Rui Li, Shen Zhao, Xiaodan Liang

Actions specify what a robot should execute, but do not describe what those commands will achieve. Vision-language-action (VLA) policies nevertheless learn primarily by predicting action chunks from images, language, and state.

VLA灵巧操作
阅读论文全文（PDF）
arXiv
VLA专业

Splat2Act: From Captured Scenes to Personalized VLA Manipulation

Xiaomeng Ye, Krishna Jaganathan, Gyubeum Lim, Gunjan Chhablani, Muhammad Zubair Irshad, Zsolt Kira

A user who brings a generalist manipulation policy into their home or office wants it to pick up their objects from their shelves, counters, and tables. Yet even a strong generalist like π_(0.5) achieves 0% pickup success in our captured deployment scenes without adaptation, and collecting real demonstrations in every user's environment is prohibitively expensive.

VLA
阅读论文全文（PDF）
arXiv
VLA专业

LAPT-Scale: Scaling Embodied Policies with Latent Action Pre-Training on Human Videos

Zhiyuan Xiang, Peidong Liu, Mingyang Li, Jiawei Li, Wenhow Li, Wenbo Ding, Xiang Deng

Scaling generalist robotic policies is limited by the high cost of collecting embodied interaction data. Large-scale egocentric human videos offer a rich alternative, but they lack robot action labels, and existing action annotation methods are often noisy under real-world visual complexity and diverse manipulation dynamics.

VLA
阅读论文全文（PDF）
arXiv
World-action model专业

世界动作模型需要什么视觉表征：银河通用与中科院自动化所的系统性对比

Kai Liu, Jiangran Lyu, Tingyu Deng, Yixin Zheng, Jingtian Wan, Wenbo Cui, Mi Yan, Yizhou Zhou, Haoran Li, Dongbin Zhao, Zhizheng Zhang, He WangInstitute of Automation, Chinese Academy of Sciences; Galbot; Peking University; University of Science and Technology of China

银河通用（Galbot）与中国科学院自动化研究所等机构发布预印本，系统比较了哪些现成图像编码器能让基于视频生成的机器人大脑更可靠，结论是一个强大的通用编码器贡献了大部分提升。据此得到的方案在人形机器人桌面任务基准RoboCasa GR1上以84.7%刷新纪录，并在银河通用实体机器人的三项任务上把成功率各提高10个百分点。

世界模型VLA机器人基础模型模仿学习
GalbotNVIDIAPhysical Intelligence
阅读论文全文（PDF）
arXiv
Tactile VLA专业

SimpleTouch 无需触觉策略预训练，50条演示在 UniVTAC 六任务平均77.5%

Chen Yang, Linzhe Shi, Changjie Wu, Hang Zhang, Ronghan Chen, Lingjun Zhang, Xu Hu, Mu Xu, Jiansheng Fan, Chen Wang

SimpleTouch 是一篇 arXiv 预印本。它给 π0.5 增加一个触觉专家，读取冻结 T3 编码器的全部 token，只用任务演示和未来触觉预测监督训练，无需触觉策略预训练或单独对齐。在六项 UniVTAC 任务上平均成功率77.5%，比 FTP-1 高10.8个百分点；在四项真机任务上平均71.3%，比 FTP-1 高8.8个百分点。

VLA触觉传感
阅读论文全文（PDF）
arXiv
VLA专业

FastOPD：面向轻量级VLA部署的在线策略蒸馏框架

Yoojin Oh, Jeongsol Kim, Yeonwoo Seo, Jangho Park, Seonghyun Jin, Sunwoo Park, Youngmin Kim, Youngjun Jun, Kyumin Choi, Jong Chul Ye

FastOPD是一个预印本框架，通过在线策略蒸馏将大型VLA模型压缩为轻量级少步学生模型。在LIBERO上，仅用2步推理就保留了π0.5约84%的性能，并将推理延迟降低78.1%；在RoboTwin 2.0上单步成功率提升15.9个百分点。该工作大幅降低训练和推理成本，有望推动基础策略在消费级硬件上的部署。

VLA
Foundation
阅读论文全文（PDF）
arXiv
Mobile manipulation专业

MobiAgent 双闭环框架以可组合原子技能实现移动操作自我提升，BEHAVIOR-1K 成功率 65.0% 超 π0.5-TA 22.5 个百分点

Chenzhi Liu, Yue Zhang, Jiehong Lin, Jianan Wang, Bo Wang, Zhongrui Wang, Xiaojuan Qi

MobiAgent 是一项预印本研究，提出双闭环智能体框架，把高层推理与低层控制解耦为可组合原子技能，内环负责部署执行与视觉反思，外环自动筛选部署数据并持续微调技能。在 BEHAVIOR-1K 仿真基准上，MobiAgent 平均成功率达到 65.0%，比任务专用基线 π0.5-TA 高 22.5 个百分点；在真机 Astribot S1 上，三轮自我改进后成功率从 32.5% 提升到 57.5%。

VLA
阅读论文全文（PDF）
arXiv
VLA Defense专业

Detect and Suppress 通过 SAE 特征抑制对抗补丁提升 VLA 鲁棒性

Yukiya Horiba, Koshiro Aoki, Shunsuke Yasuki, Bum Jun Kim, Taiki Miyanishi

这是一篇 arXiv 预印本。研究者用稀疏自编码器分析 VLA 内部表征，找到与对抗补丁高度相关的特征，并在推理时仅当线性探针检测到攻击时抑制该特征。在 LIBERO-10 上，该方法在间歇攻击下提高成功率，例如 π0.5 在 n=10 攻击下成功率从 26.0% 升至 32.4%，而连续抑制会严重损害性能，说明条件干预的重要性。

VLA
阅读论文全文（PDF）
arXiv
VLA manipulation专业

ChunkVLA-AM 将 OpenVLA-OFT 部署至 FAIRINO FR3，增材制造抓取真机成功率 92.9%

Zhugang Liu, Kaichuang Zhang, Jinman Zhang, Pu Sun, Martha Asare, Jose Hernandez, Maxim Ermolinsky, Efren Saenz, Qi Lu, Jinghao Yang

该预印本提出 ChunkVLA-AM 系统，在 FAIRINO FR3 机器人上部署 OpenVLA-OFT，实现增材制造工作单元内的语言指令物体转移。在 42 次真机 A 到 B 抓取试验中成功 39 次，成功率 92.9%，且照明鲁棒性分析给出 85 至 125 的低误差亮度区间。这为将视觉语言动作模型引入增材制造后处理提供了可复现的部署流程。

VLA
阅读论文全文（PDF）
arXiv
Unified world-action model专业

UniWAM统一世界-动作模型：融合物理推理、视觉生成与动作预测，LIBERO与RoboTwin刷新SOTA

Jiayi Chen, Wenxuan Song, Jingbo Wang, Shuai Zhou, Xicheng Gong, Zehua Fan, Ziyang Zhou, Junwu E, Haodong Yan, Fuhao Li, Qize Yu, Xu Huang, Pengwei Wang, Wen Chen, Shunbo Zhou, Haoang Li

UniWAM是一个统一世界-动作模型，将物理推理、视觉生成和动作预测集成在单一混合Transformer架构中，并在约1万小时人类与机器人数据及829万VQA样本上预训练。该预印本在LIBERO、RoboTwin等基准上取得SOTA，真机实验指令跟随与长程任务也优于π0.5和Motus。其重要之处在于展示了人类-机器人联合预训练的对数线性扩展规律，为通用机器人模型提供数据与架构路径。

VLA
阅读论文全文（PDF）
arXiv
Manipulation专业

DynaHarness动态物理执行框架将LIBERO-Pro成功率从17.5%提升至75.2%

Haoyuan Deng, Jiebin Liu, Tengxiao Zhang, Langning Yan, Hongye Cao, Ziwei Wang

本文为预印本。DynaHarness通过动态物理执行合同连接慢速语义推理和快速物理治理，在LIBERO-Pro新增800个状态上取得75.2%成功率，冻结π0.5策略仅17.5%。同一技能库下，完整动态执行74.0%，对比一次性重规划63.9%，显示对已有能力进行实时监控、替换和恢复的价值。

VLA
阅读论文全文（PDF）
arXiv
Safe VLA专业

FailBank：将运行时反馈变为 VLA 策略的持久自我进化

Mingyue Cui, Zheyuan Liu, Yihan Zhu, Zheyuan Zhang, Meng Jiang

FailBank 是一个四阶段自进化框架，将运行时 CBF 屏蔽的修正建议转化为学习记录，用于持续更新视觉-语言-动作模型，而无需在部署时使用屏蔽。在 VLA-Arena 静态障碍套件上，FailBank 在两个骨干模型上分别将任务成功率提高 8.5 和 6.9 个百分点，同时将策略诱导累积成本降低 35.6% 和 23.8%。与运行时屏蔽 AEGIS 相比，成功率分别提高 25.4 和 9.5 个百分点。本文为预印本，尚未经过同行评审。

VLA
阅读论文全文（PDF）

主题页

操作 8VLA 4World Action Models 5World-action model 3Aerial manipulation 3Dexterous manipulation 3Sim-to-Real Manipulation 3Social Navigation 3

论文来自 arXiv 机器人领域。尚未生成摘要的论文，显示摘要开头部分。