🤖 具身智能行业新闻

Robust robotic manipulation in the real world requires not only an understanding of the current observation, but also memory and dynamics modeling. World action models (WAMs) possess these capabilitie...

提出一种带持久记忆的世界动作模型（WAM），不仅理解当前观测，还能进行动力学建模和长期记忆，提升真实世界机器人操作鲁棒性。

Robot learning has advanced rapidly in learning control, but learning the physical body of a robot remains much more difficult because jointly searching over design and control creates a very large co...

探讨如何将人类操作视频转化为多指机器人手的控制指令，并联合优化机器人的物理形态设计（例如手指长度）和控制策略。

Embodied foundation models are expected to benefit from data scaling like large language models, but face a much tighter data bottleneck. Teleoperated real-robot trajectories remain the dominant pretr...

指出遥操作等真机数据存在严重瓶颈，研究发现利用大规模第一视角的人类视频数据进行具身大模型的预训练，效果竟能超越真机采集的数据。

This paper presents an experimental study of motion planning for resilient continuum robots. In this study we mainly focused on multi-criteria decision-making, its application for path-planning algori...

本文针对具有高弹性的连续体机器人（一种由柔性材料构成、可无限弯曲的机器人）进行了运动规划实验研究，重点探讨了多准则决策在路径规划算法中的应用。

Vision-based tactile sensors require high-fidelity simulation for reinforcement learning, yet existing approaches struggle to provide accurate mechanical stress fields within GPU-accelerated robotics ...

视觉触觉传感器需要高保真仿真来训练强化学习模型，但现有方法难以在GPU加速的机器人模拟中提供准确的机械应力场。该研究提出了基于有限元方法（FEM）的可扩展框架TaCauchy来解决此瓶颈。

Continuum robots offer strong potential for manipulation tasks due to their high degrees of freedom, compliant structures, and operational safety. However, their adoption in both research and practica...

连续体机器人凭借高自由度和柔顺结构在操作任务中潜力巨大，但在研究与应用中缺乏统一的软硬件标准。该研究提出了CoLI平台，结合一体化3D打印技术与同构遥操作系统，降低了数据采集与算法验证的门槛。

The ability to localise teams of robots is essential for applications ranging from robotic fleets in unstructured environments to cooperative control and navigation tasks. In such contexts, fixed infr...

仅利用测距测量实现多机器人编队的相对定位，无需依赖外部固定基础设施（如GPS或UWB阵列）。

Vision-language-action (VLA) models show strong capabilities in single and dual-arm robotic manipulation. Prior works show coordinated bimanual behaviors can emerge from end-to-end learning, leveragin...

视觉-语言-动作（VLA）模型在机器人操作上表现出色。以往依赖端到端学习来涌现双臂协调行为，该研究提出Co-VLA，在模型架构中引入协调感知的结构化动作建模，以提升双臂协作任务的执行效率。

AI vision models are a driving factor for the potential use case scenarios of cognitive robotics within in the industry and household applications. A large array of methods from semantic environment a...

探讨如何利用AI视觉模型高效生成合成数据，弥合真实场景差距，推动认知机器人在工业和家用场景中的应用。

Vision-Language-Action (VLA) models pre-trained on massive video-robot datasets have revolutionized robotic manipulation, yet their multi-billion parameter architectures impose prohibitive computation...

在海量视频和机器人数据上预训练的视觉-语言-动作（VLA）模型具有强大的操作能力，但其数十亿参数的架构带来了巨大的计算负担。研究表明，在实际应用微调时，仅需更新较少的模型层即可达到优异效果。

Joint spatial and temporal understanding of 3D scenes is a crucial requirement for robots deployed in everyday household environments. Such agents must not only comprehend and navigate spatial layouts...

家庭环境中部署的机器人需要具备联合时空理解能力。该研究提出FlowMaps，利用流匹配（Flow Matching）技术对场景中物体的长期、多模态动态变化进行建模，帮助机器人理解复杂的环境演变。

Parallel-jaw grippers are the default manipulator choice in robotics because they are simple, robust, and inexpensive. Their limited in-hand mobility, however, often forces large arm motions and restr...

平行爪夹爪因简单耐用且便宜成为机器人标配，但其手内操作 mobility 有限，常需依赖机械臂大范围运动。该研究提出Belt-Finger，一种基于软体带驱动的平价夹爪，以较低成本实现了高灵巧度的手内重定向操作。

Human Action Recognition (HAR) is frequently investigated in Human-Robot Collaboration (HRC) research to understand what actions have been performed and hence the state of a collaborative task. Accura...

利用人体动作识别（HAR）来精确判断协作任务的进度和状态，提升人机协作装配过程的准确性与鲁棒性。

In Shenzhen, workers at IO-AI Tech control humanoid robots using a VR rig reminiscent of Ready Player One.

在深圳，IO-AI科技公司的员工通过类似于《头号玩家》中的VR设备，对 humanoid 机器人进行遥操作控制并收集数据。

We present a zero-shot framework for long-horizon dexterous manipulation that grounds language instructions into executable 3D task plans from calibrated multi-view RGB images. Rather than training an...

提出了一种零样本长时程灵巧操作框架，能将标定好的多视角RGB图像中的语言指令转化为可执行的3D任务计划，无需额外训练。

How can we scalably generate data for robotic manipulation, especially on human-like platforms such as dexterous multi-fingered hands? Learning from human videos has recently emerged as a likely answe...

探讨如何从日常人类视频中 scalable 地生成数据，以训练多指灵巧手等类人机器人平台的操作能力。

World models are the next big thing in AI beyond LLMs and, with this round, Odyssey has cemented itself as one of the startups to watch.

主打世界模型的初创企业 Odyssey 宣布获得新一轮融资，估值达14.5亿美元，投资方包括亚马逊等科技巨头。

This study presents a method for modeling diverse plant branches by iteratively estimating material parameters to support delicate branch manipulation. Branch manipulation is necessary in agricultural...

提出一种通过迭代估计材料参数来对各类植物树枝进行建模的方法，以支持农业机器人在修剪等场景下的精细柔顺操作。

Embodied Vision-Language-Action (VLA) models are typically obtained by fine-tuning powerful pretrained VLMs on robotics data, yet it is unclear how much commonsense and factual knowledge they retain a...

探究通过对强大预训练VLM进行机器人数据微调得到的VLA模型，究竟保留了多少关于真实世界的常识和事实知识。

Continuum robots offer a promising approach for minimally invasive and natural-orifice surgical procedures due to their inherent compliance and dexterity. However, this flexibility also makes estimati...

针对连续体机器人在微创手术中难以估计形状的问题，提出利用飞秒激光直写技术进行高精度形状感知的方法。

Markerless, single-RGB-D-camera motion capture provides a low-cost and non-invasive alternative to conventional marker-based systems for robot teleoperation; however, depth estimation often degrades i...

提出一种单RGB-D相机的无标记动作捕捉方案，通过确定性的手臂运动学校正解决深度估计退化导致的遥操作遮挡问题。

In this letter, we present a hierarchical control framework that enables wheeled bipedal robots to perform planar object sliding tasks with their wheeled legs. The proposed approach formulates a nonli...

提出一种分层控制框架，使轮式双足机器人能够利用其轮腿执行平面物体滑动任务，结合非线性规划和阻抗控制。

This paper presents an invertible neural network adapter for general robotic manipulation, designed to generate precise high-dimensional actions conditioned on multimodal observations, including visua...

提出一种可逆神经网络适配器，基于多模态视觉观测生成精确的高维动作指令，提升通用机器人操作中的流匹配效率。

提出一种能够根据自然语言指令，准确预测和生成高维3D人体或机器人运动轨迹的新模型。

Establishing a universal benchmark for tactile representation learning in robotic manipulation remains challenging due to the diversity of tactile sensor designs, data formats, and robot embodiments. ...

针对触觉传感器设计差异大的问题，建立了一个通用的全手触觉表征学习基准，结合第一视角视觉进行评测。

If physical AI is going to match the accomplishments of LLMs, there's a data problem that needs to be solved.

指出物理AI要匹配LLM的成就，必须先解决数据瓶颈问题，目前已有实验室向XDOF等公司付费购买或定制真实的机器人操作数据。

Dynamic 3D hand reconstruction from egocentric videos is essential for next-generation computing platforms such as AR/VR and AI glasses. Despite its importance, most prior works focus either on multi-...

提出一种前馈3D高斯溅射（3DGS）方法，从第一视角视频直接进行动态4D手部重建，对AR/VR和AI眼镜至关重要。

Autonomous robots operating under forest canopies need robust perception of trees and surrounding vegetation across varying seasonal conditions. Existing forestry datasets provide lidar or camera data...

发布了一个包含激光雷达、雷达和相机的多模态森林场景数据集，以提升自动驾驶机器人在不同季节下对树木的感知鲁棒性。

Sidewalks in the real world are crowded, cluttered, and less structured than roads, making 3D occupancy prediction a key ingredient for the safe navigation of mobile robots such as delivery bots and e...

针对真实人行道拥挤杂乱的特点，提出一种基于单目相机的2D-3D混合学习方案，用于配送机器人等安全导航所需的3D占据预测。

Task-oriented grasping performance degrades significantly when object views suffer from occlusions. Existing task-oriented grasping methods typically assume task-relevant regions are visible in the in...

针对物体视角被遮挡导致任务导向抓取性能下降的问题，提出一种基于可供性引导的主动视图规划方法。

Vision-Language Models (VLMs) enable robots to follow open-language instructions. However, dense VLM embeddings have shown to be noisy and lack spatial consistency. This is problematic for robotic app...

针对视觉语言模型（VLM）的稠密特征嵌入缺乏空间一致性的问题，提出一种提升机器人任务中空间语义连贯性的方法。

Visual servoing with self-supervised Vision Transformer (ViT) features enables training-free robotic positioning with strong generalization, but faces a fundamental trade-off between robustness and pr...

利用自监督ViT特征实现免训练机器人定位，通过自适应分辨率分块技术（ART）解决鲁棒性与精度之间的权衡问题。

Autonomous navigation of quadrupedal robots in diverse environments fundamentally relies on resilient Simultaneous Localization and Mapping (SLAM). While visual-inertial SLAM has matured across wheele...

系统评估了四足机器人在不同地形下，不同传感器配置对多模态同步定位与建图（SLAM）鲁棒性的影响。

Autonomous mobile service robots are often required to complete tours that require navigating through a set of locations in an environment. Example domains include guiding people through a shopping ma...

提出一种拥堵感知的路径规划方法，帮助自主移动服务机器人在商场、博物馆等拥挤环境中完成多点导航巡视任务。

On April 19, 2026, theHonor Lightning humanoid robot ran a half-marathon in 50 minutes and 26 seconds, beating the human world record by 7 minutes and the best rob

回顾荣耀闪电人形机器人在半程马拉松中以50分26秒的成绩打破人类世界纪录，分析其背后的硬件和算法优势。

Action-conditioned world models have emerged as a promising paradigm for robot learning, offering a scalable alternative to costly real-world experimentation by generating action-consistent video roll...

提出一种记忆增强的动作条件世界模型，通过生成动作一致的视频轨迹提供长期记忆，解决长时间任务中的状态跟踪问题。

Tactile sensing provides direct measurements of contact interactions that are essential for robotic manipulation. However, current simulators lack the fidelity to faithfully model the complex deformat...

提出一种在仿真与现实之间建立物理增强的共享潜空间方法（TactSpace），以解决触觉传感器在仿真器中形变建模不准确导致Sim-to-Real困难的问题。

介绍如何利用 Hugging Face 开源社区中的模型和 LeRobot 框架，结合 Strands Agents 快速搭建并驱动真实的机器人硬件。

The next humanoid robot might not have a head. It might not have legs. It might even sit on a wheeled base and fold down like a deck chair. But, as Genesis AI puts it, "humanoid robots don't need to l...

探讨未来人形机器人的形态演变，认为可能不再有头部和双腿，而是采用轮式底座或折叠形态，只要能完成任务即可。

Effective human-robot teamwork requires robots to adapt to partners, situations, and task dynamics from the start of an interaction. In the MATRX Urban Search and Rescue (USAR) environment, people can...

在MATRX城市搜救（USAR）环境中，研究如何利用情景记忆让机器人记住并适应之前的人类协作模式，实现开箱即用的高效人机协同。

Robots deployed in the real world should learn from their experience and improve over time. This requires a mechanism of practicing and learning from feedback. In this paper, we propose VERITAS, a gen...

提出VERITAS框架，让真实世界部署的机器人能够通过视觉反馈机制从经验中学习并持续改进操作策略。

Collaborative human-object interaction shows dynamic and complex movements that require mutual anticipation and continuous adjustment between participants and the shared object. Modeling such collabor...

提出一种针对人、物体协同交互的动态复杂运动建模方法，能够更好地预测和调整多方协作时的连续动作。

We present EBench, a simulation benchmark that diagnoses generalist mobile manipulation policies beyond a single success-rate scalar. EBench comprises 26 diverse and challenging manipulation tasks ann...

提出EBench仿真测试基准，不再仅用单一成功率来衡量移动操作机器人，而是通过26个任务进行多维度的能力诊断。

Zero-Shot Object-Goal Navigation (ZS-OGN) requires embodied agents to explore and locate target objects without any prior training. To this end, recent methods leverage foundation models. But they typ...

零样本目标导航要求机器人在无训练数据的情况下寻找目标。EvolveNav 结合大模型的主动预反思和自我进化记忆机制，突破了现有基础模型在此任务上的局限。

The Israeli tech company and Intel subsidiary said it will launch its own robotaxi service in a U.S. city in 2027.

这家以色列科技公司及英特尔子公司宣布，将于 2027 年在美国某城市推出自己的无人驾驶出租车服务，从而同时涉足自动驾驶的软硬件供应与终端运营。

Current world models face a fundamental tension: faithful long-horizon simulation demands deep computation, but deeper models are expensive to deploy and prone to compounding errors. We resolve this b...

为解决现有世界模型在长时间模拟中面临的计算成本高和误差累积问题，提出一种通过深层循环计算来优化模拟的框架。

Conventional human-in-the-loop approaches typically involve users only when a robot encounters failure or uncertainty, treating humans primarily as tools for improving robot performance. However, in m...

打破仅在人机交互失败时才引入人类干预的传统模式，提出一种基于人类参与度感知的人机在环新框架。

Finite-dimensional (FD) diffusion policies exhibit temporal drift owing to discretization artifacts that degrade long-horizon performance (when deployed on physical systems). We introduce a backward K...

为解决有限维扩散策略因离散化导致的时间漂移和长时间序列性能下降问题，引入向后Kolmogorov方程进行优化。

We break down Qwen-RobotSuite, the Qwen team's three new embodied AI models. We cover RobotManip, a Vision-Language-Action model built on Qwen3.5-4B for manipulation. We cover RobotWorld, a language-c...

阿里 Qwen 团队推出包含三个具身智能模型的 Qwen-RobotSuite：基于 Qwen3.5-4B 的操作模型 RobotManip、语言条件视频世界模型 RobotWorld 以及导航模型。

Tired of vacuuming? Hand the reins to a robot vacuum.

盘点 2026 年市面上的主流扫地机器人产品，包括 Shark 和 Eufy 等品牌，为消费者提供自动化清洁家电的选购指南。

Agentic navigation systems require a base navigation model whose observation strategy can be externally reconfigured at inference time, because instruction following, object search, target tracking, a...

智能体导航系统要求基础模型能够在推理时动态更改观测策略，以应对指令跟随、目标搜索和目标追踪等不同任务。Qwen 团队发布了 Qwen-RobotNav 以解决该问题。

I have one golden rule: It's not about the mower, it's about the yard. Use my free checklist to decide which model to buy.

作者基于大量割草机器人的使用经验总结出一条黄金法则：重点不在机器本身，而在于庭院的地形环境，并提供了一份选购清单。

Deformable Linear Objects (DLOs), such as wires and cables, are central to industrial assembly. Unlike rigid objects, whose state is captured by a 6-DoF pose, DLOs have an infinite-dimensional configu...

针对工业装配中常见的柔性线缆（DLOs）无限维状态难以建模的问题，推出了专门的仿真测试基准。

Cross-end-effector grasp generation seeks a unified model that generalizes across objects and across embodiments ranging from parallel grippers to dexterous end effectors. Existing grasp generators ar...

旨在解决不同机器人末端执行器（如平行夹爪或灵巧手）之间的抓取泛化问题，通过几何感知实现对齐。

We propose a hybrid grasp synthesis framework that combines a learning-based Energy-Based Model (EBM) with an analytical Iterative Closest Point (ICP) method to generate robust grasps from partially o...

结合基于学习的能量模型（EBM）和分析学方法的迭代最近点（ICP）算法，在遮挡等部分观测条件下生成稳定的抓取姿态。

Vision-language-action models (VLAs) combine vision-language backbones with expressive generative action heads trained via flow matching on large-scale robotic datasets. Despite their strong empirical...

针对目前主流的通过流匹配训练的VLA模型，提出量化其动作输出不确定性的方法，以提升模型部署的安全性。

Diffusion-based visuomotor policies deployed with asynchronous inference often exhibit inter-chunk discontinuities and lack explicit mechanisms for obstacle-aware execution, leading to jerky motions a...

解决基于扩散模型的视觉运动策略在异步推理时产生的动作不连贯和缺乏避障机制的问题，实现平滑操作。

The service will leverage its Moovit platform to launch in an a US city in 2027.

Mobileye 计划于 2027 年在美国某城市推出其无人驾驶出租车服务，该服务将依托其旗下的 Moovit 出行平台进行运营。

Most Vision-Language-Action (VLA) models map observations directly to actions without explicit reasoning, limiting their capacity for reasoning-intensive long-horizon tasks. To address this, existing ...

现有的视觉-语言-动作（VLA）模型大多将观测直接映射为动作，缺乏显式推理。ThinkingVLA 引入视觉与语言交织的推理机制，以解决长视野复杂任务。

Current Vision-Language-Action (VLA) models face a trade-off between efficient action generation and explicit deliberation. Directly decoding actions from vision-language backbone representations enab...

解决了当前VLA模型在快速动作生成和显式逻辑推理之间的权衡问题，通过在潜在空间逐步优化动作计划来提升表现。

Recent World-Action (WA) models demonstrate strong generalization ability and data efficiency, but they typically rely on expert trajectories for training. This reliance limits their ability to acquir...

打破了世界-动作模型高度依赖专家轨迹的限制，引入强化学习与重建奖励，使其能从在线视频中学习并探索。

Genesis AI, the French robotics startup backed by former Google CEO Eric Schmidt, unveiled its first general-purpose robot ​on Tuesday, as AI capabilities expand beyond chatbots and into physical ‌mac...

由前谷歌 CEO 埃里克·施密特支持的法国机器人初创公司 Genesis AI 于周二发布了其首款通用机器人。随着 AI 能力从聊天机器人向实体物理机器扩展，该公司选择了一条与人形机器人不同的技术路线。

Foundation models in language and multimodality achieve strong generalization by aligning heterogeneous data under a unified formulation and training at scale. In this report, we investigate whether t...

Qwen 团队探讨如何通过统一框架对齐异构数据并进行大规模训练，以使机器人操作基础模型像语言和多模态模型一样获得强大的泛化能力。

Humanoid robots promise whole-body interaction in human-centered environments, but scalable policy learning remains difficult because task-level decision-making and whole-body dynamic execution are ti...

人形机器人有望在人类环境中实现全身交互，但任务级决策与全身动态执行的分离导致策略学习困难。HumanoidArena 旨在对这种分层全身学习进行基准测试。

Robotics are expected to support environmental monitoring and natural disaster management, where decisions must be made under uncertainty, resource limitations, and strict operational constraints. In ...

针对环境监测和灾害管理中资源受限和需在不确定性下决策的难题，研究提出 ED3R，利用多机器人协作智能体进行具备能耗感知的分布式灾害检测。

Generalist embodied agents require more than object recognition: they must reason about spatial relations, actions, procedures, human intentions, environmental constraints, and commonsense consequence...

强调通用具身智能体不能仅靠物体识别，还需要对空间、意图、环境约束和常识后果进行复杂推理，并推出相应评测基准。

Safe and efficient trajectory planning in unknown, cluttered 3D environments constitutes a critical bottleneck for deploying Unmanned Aerial Vehicles (UAVs) in real-world applications. This challenge ...

解决无人机在未知、杂乱的3D环境中进行安全高效轨迹规划的瓶颈，提出无需先验地图的视场角受限主动感知方法。

Humans naturally leverage diverse sensing modalities to interact with the physical world, while most Vision-Language-Action (VLA) models for robotics rely solely on RGB observations. This limits their...

打破当前机器人VLA模型过度依赖RGB视觉的局限，像人类一样引入多种传感模态（如触觉、深度）进行物理交互。

Simultaneous localization and mapping using radar sensors has gained increasing attention due to radar's inherent robustness to adverse weather and lighting conditions. However, radar measurements are...

针对雷达在恶劣天气和光照下鲁棒但存在测量噪声的问题，提出一种新的雷达同步定位与建图（SLAM）方法。

Robot mowers aren't one-size-fits-all devices - many have overhyped specs you don't even need. Here's my advice after years of testing.

割草机器人并非一刀切的设备，许多产品存在过度宣传的参数。作者基于多年测试经验指出，消费者应警惕无用的营销指标，关注实际场景需求。

The ability to react dynamically to tactile signals has long been considered crucial to agile human-level dexterity. Yet contemporary learning-based Vision-Language-Action (VLA) models for robotic man...

针对当前基于视觉-语言-动作（VLA）模型的机器人难以对触觉信号做出动态反应的问题，研究提出 T-Rex，旨在提升机器人实现人类级别敏捷操作的触觉反应能力。

Humans can grasp objects effortlessly, whereas multi-fingered robots are far from this level of generality. We argue that the most natural source of robot grasping data is from humans, who pick up tho...

为了让多指机器人达到人类级别的通用抓取能力，研究主张利用人类抓取物体的自然动作作为最核心的数据源来训练机器人抓取策略。

Generalist robot policies must follow user instructions while reasoning about how objects, cameras, and robot actions interact in the 3D physical world. Recent vision-language-action models (VLAs) and...

为了让通用机器人策略在遵循用户指令的同时理解三维物理世界的交互，研究引入几何动作模型，改进现有的视觉-语言-动作模型（VLA）在空间推理上的不足。

When pretrained VLA policies are fine-tuned through online RL, each rollout episode produces only a single binary outcome (success or failure), yet the actor update requires per-transition supervision...

针对 VLA 策略在在线强化学习微调时仅能获得成功或失败这种单一稀疏二值奖励的问题，研究提出一种分层优势加权方法，为每步状态转移提供细粒度监督信号。

Spatial generalization is critical for imitation-learned manipulation policies, but achieving it typically requires scaling demonstrations across diverse object poses, robot configurations, and camera...

模仿学习策略对空间泛化要求极高，通常需要海量不同位姿和视角的演示数据。R2RDreamer 利用 3D 感知技术进行数据增强，以低成本解决 2D 操作策略的空间泛化难题。

We introduce Qwen-RobotWorld, a language-conditioned video world model for embodied intelligence. With natural language as a unified action interface, it predicts physically grounded future visual tra...

Qwen 团队发布 Qwen-RobotWorld，这是一个以自然语言为统一动作接口的具身智能视频世界模型，能够预测具有物理规律的未来视觉轨迹。

Human interventions provide crucial corrective signals for post-training Vision-Language-Action (VLA) models. However, enabling seamless humanoid interventions is a formidable systems challenge due to...

人类干预能为 VLA 模型的后训练提供重要的纠错信号，但实现无缝的人形机器人干预在系统架构上极具挑战。ROVE 利用强化学习解决了这一难题。

Cyber-physical system (CPS) controllers are vulnerable to faults and malicious attacks, including failures triggered only under complex plant conditions, yet pre-deployment validation typically relies...

信息物理系统（CPS）控制器易受复杂故障和恶意攻击影响。研究提出引入沙盒机制的数字孪生技术，以弥补现有部署前验证方法的不足，提升系统安全性。

For residual learning that refines existing behavior, sample efficiency depends on two things: how much information each rollout returns, and how efficiently the learner uses that information. Reinfor...

研究探讨了如何提高残差学习的样本效率，通过强化学习优化真机五球抛接任务，重点分析了每次采样获取的信息量及其利用效率。

Robots operating in non-stationary environments must continually adapt their policies as the dynamics drift, but onboard energy and compute budgets cap how often a full state estimation and re-plannin...

针对非平稳环境中机器人重规划受限于机载能源和算力的问题，研究提出一种基于“遗憾值”的引导机制，动态决定机器人何时进行状态估计和重新规划。

Urban sidewalk navigation presents significant challenges due to complex structural layouts, dynamic pedestrian behaviors, and long distances. While recent visual navigation models offer a promising s...

城市人行道导航因复杂的结构布局、行人动态和长距离面临巨大挑战。研究推出 SidewalkBench，用于评估和基准测试最新的视觉导航模型在城市无障碍环境中的表现。

Rovers rely on perception to maintain spatial maps that encode both objects and sensor quality (e.g., range reliability, lighting artifacts, data density), guiding data fusion, embedding updates, and ...

该框架提出一种置信度感知的开放词汇语义地图。火星车等机器人利用该地图编码物体属性和传感器质量（如测距可靠性、光照伪影），从而指导更精准的数据融合与导航决策。

This paper addresses the problem of learning an output-feedback surrogate controller offline that approximates a given, possibly computationally expensive, nonlinear controller-observer pair. The surr...

研究提出一种离线学习输出反馈代理控制器的方法，用以替代计算成本高昂的传统非线性控制器-观测器对，从而在保证稳定性的同时降低在线计算开销。

Wireless communication channel characterization for unmanned aerial vehicles (UAVs) is essential for reliable control, data transmission, and mission performance in civil, industrial, and defence appl...

针对无人机（UAV）在民用、工业和国防应用中的可靠控制与数据传输需求，研究对其无线通信信道的特性进行了分析与建模。

We present Unified Motion-Action (UMA) Model, an approach that uses 3D object motion trajectories as a shared interface to bridge visuomotor control and dynamics modeling. UMA treats object motion and...

研究提出统一动作-运动（UMA）模型，利用3D物体运动轨迹作为共享接口，打通了视觉运动控制与动力学建模，使异构机器人能共享学习到的动作策略。

This work addresses spatial question answering for service robots traversing long egocentric routes. Given a query such as "where can I find a dry cleaner on the way back home?", the system returns a ...

该研究解决服务机器人在长距离第一视角路径上的空间问答问题（如“回家路上哪里有干洗店”），系统利用视觉语言模型（VLM）进行空间目标追踪与导航。

Detecting unanswerable user queries remains essential for the reliable deployment of real-world embodied agents. However, modern vision-language models (VLMs) often generate overly confident answers e...

为了可靠部署具身智能体，检测无法回答的用户查询至关重要。研究提出 Semantic Flip 方法，合成分布外（OOD）数据，解决视觉语言模型（VLM）在面对无法回答问题时过度自信的幻觉问题。

Learning from Demonstration (LfD) enables intuitive robot skill acquisition by allowing robots to learn directly from human task demonstrations. However, current methods often fail to address the fact...

从人类示范中学习（LfD）让机器人直观获取技能，但现有方法常忽略人类示范包含错误或非最优动作的问题。LOPAL 提出局部性能感知的主动学习，识别并修正这些不完美的示范。

We introduce a data-efficient distributed Simultaneous Localization and Mapping (SLAM) framework designed for a team of robots equipped with LiDAR, cameras, and inertial sensors. Our framework uses sc...

提出一种数据高效的分布式同步定位与建图（SLAM）框架，供配备激光雷达、相机和惯性传感器的多机器人团队使用，通过场景图匹配减少通信带宽需求并提升建图效率。

Adaptive torque prediction in dynamic exoskeleton scenarios requires expensive motion capture systems, which are infeasible in complex outdoor environments. Trajectory prediction has emerged as one of...

动态外骨骼场景下的自适应力矩预测通常依赖昂贵的动作捕捉系统，这在复杂户外环境中不可行。该研究提出基于轨迹预测的方法，实现适应复杂环境的通用下肢外骨骼辅助策略。

Realistic visual simulation of food manipulation requires accurate material parameters, yet these are difficult to measure directly and vary across the heterogeneous regions of a single food item. We ...

食物拟真视觉模拟需要准确的材料参数，但这些参数难以直接测量。研究提出在潜空间中使用强化学习方法来估计食物的异构材料参数。

Generalist manipulation policies are increasingly presented as foundation models for robotic control, but their real-world generalization remains difficult to diagnose. A policy may succeed on demonst...

通用操作策略常被视为机器人控制的基础模型，但其真实世界泛化能力难以评估。ATOM-Bench 提供了一个真实世界基准，诊断策略在原子技能及长程组合任务上的泛化表现。

Foundation models are reshaping robotics by enabling robots to interpret open-ended instructions, reason over multimodal contexts, and operate in complex, open-world environments. However, their integ...

基础模型让机器人能理解开放式指令并在复杂环境中运行，但整合这些模型也带来了严重的安全与隐私漏洞。本文系统梳理了相关攻击面及防御机制。

Exploration and object search require robots to perceive their environment, identify regions of interest, and plan trajectories that improve target-detection likelihood or maximize information gain. M...

机器人在执行探索和目标搜索时，需要识别感兴趣区域并规划能最大化信息增益的轨迹。DIFF-IPPO 结合扩散模型与开放词表地图，有效提升了复杂环境下的路径规划质量。

Generalist robot policies require trustworthy evaluation and robot-usable training data, but both are difficult to scale with physical robots alone. Real-robot trials and demonstrations remain the mos...

通用机器人策略需要大量高质量的训练数据和可靠的评估，但纯靠真实机器人成本极高。DataLadder 提出了一套连接仿真与真实数据的互转工具链，以高效扩充具身智能训练数据。

Achieving expert-level expressive full-body motion tracking across multiple humanoids solely from demonstration data remains a challenging and relatively an underexplored problem in humanoid robot lea...

仅从示范数据中学习专家级的全身表达性运动追踪，在仿人机器人领域仍具挑战。VENOM 网络提出一种通用方法，解决多仿人机器人的全身精细运动追踪问题。

Learning-based manipulation policies have made substantial progress in real-world robot manipulation, particularly for short-horizon action generation. However, deployment in open workspaces remains f...

基于学习的操作策略在真实世界短视界任务中取得进展，但在开放工作空间中常因环境变化导致不可预测的失败。PATCH 提出一种潜在补丁监控方法，实时检测并防止策略执行偏离预期。

Aerial manipulators enable physical interaction in hard-to-reach environments; however, the combined problem of direct whole-body aerial manipulation under rapid arm motion, payload changes, and relat...

空中操作机器人能够在难以到达的环境中进行物理交互，但在机械臂快速运动和载荷变化下，全身控制面临巨大挑战。该研究将强化学习与内循环动力学估计器结合，提升了复杂扰动下的飞行抓取稳定性。

World models are widely used in robotic and agentic engineering control systems due to their ability to learn latent dynamics for planning and decision-making. As these systems are increasingly deploy...

世界模型因能学习潜在动力学用于规划，被广泛应用于机器人和智能体控制。随着部署规模扩大，ARB4WM 基准专门评估这些世界模型在面对对抗性干扰时的鲁棒性表现。

Expressive co-speech gestures are crucial for natural human-robot interaction, but generating them on physical humanoid robots is difficult because gesture strokes must align with speech emphasis whil...

富有表现力的语音伴随手势对自然人机交互至关重要，但在实体仿人机器人上生成这些手势极其困难，因为动作节拍必须与语音重音精准对齐，同时还要满足物理约束。WaveSync 提出了一种约束波前优化方法解决此问题。

Diffusion and flow-based generative policies provide a powerful policy class for reinforcement learning by inducing rich stochastic exploration through iterative action generation. However, the stocha...

扩散和基于流的生成策略通过迭代动作生成引入丰富的随机探索，增强了强化学习的探索能力。但这种随机性会导致控制不稳定。本研究提出一种方法，将生成式策略引导为稳定的机器控制策略。

Registering two captures of the same indoor space taken at different times underpins persistent spatial memory for robots and AR systems, yet the realistic version of this task is egocentric and its m...

对同一室内空间在不同时间点的两次捕捉进行配准，是机器人持久空间记忆的基础。PROSE 利用视觉语言模型（VLM），无需额外微调即可解决具有挑战性的第一视角场景配准任务。

Robotic systems routinely encounter conflicting objectives, modeling errors, and degenerate contact conditions that render quadratic programs (QPs) infeasible. Yet most optimization solvers and differ...

机器人系统经常遇到目标冲突、建模误差和退化接触条件，导致二次规划（QP）求解器报错不可行。Elastic ODYN 通过可微优化方法，优雅处理这些不可行约束，保证控制系统持续运行。

Humanoids deployed in human-centered environments must handle force-interactive tasks, where external contacts introduce unexpected disturbances that disrupt locomotion accuracy and stability. Existin...

部署在以人为中心环境中的人形机器人必须处理受力交互任务。外部接触引入的意外扰动会破坏行走稳定性。ADAPT 提出一种分析性扰动感知策略训练方法，提升人形机器人在外力扰动下的步态稳定性。

World models are transitioning from passive visual generators to foundational, operational infrastructure for Physical AI: they must natively acquire world knowledge from heterogeneous experience, mai...

世界模型正从被动的视觉生成器，转变为物理 AI 的基础运行架构。Kairos 提出一个原生技术栈，能够从异构经验中自主获取世界知识，并维持长期的多维一致性，用于物理系统的规划和决策。

Social robots for children with autism are often evaluated through engagement and interaction quality, assuming the robot acts as a social scaffold. We report a mixed-methods "withdrawal" study that t...

针对自闭症儿童的社交机器人通常通过互动参与度来评估。本研究报告了一项混合方法的“撤回”研究，通过挑战传统的参与度指标，重新审视社交机器人在特定群体中的真实交互体验设计。

Effective personalized AI-assisted learning demands systems that can not only generate accurate learner-specific educational materials, but also dynamically adapt their instruction to diverse learners...

有效的个性化 AI 辅助学习不仅需要生成准确的特定教材，还需根据学习者的状态动态调整教学指令。LectūraAgents 结合大模型与多智能体协同，实现教育内容生成与具身教学的动态自适应。

The integration of augmented reality (AR) and EEG-based brain-computer interfaces (BCIs) offers a promising path for enabling intuitive control of robots for assistive purposes. However, existing AR b...

增强现实（AR）与基于脑电图（EEG）的脑机接口（BCI）结合，为残障人士提供直观的机器人控制途径。该研究解决现有 AR-BCI 在通用机械臂复杂抓取任务中的局限性，提升了控制的自然度。