cs.RO（2025-09-26）

📊 共 31 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (23 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (2 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱八：物理动画 (Physics-based Animation) (1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱一：机器人控制 (Robot Control) (23 篇)

#题目一句话要点标签🔗⭐
1 Teleoperator-Aware and Safety-Critical Adaptive Nonlinear MPC for Shared Autonomy in Obstacle Avoidance of Legged Robots 提出一种遥操作感知的安全自适应非线性MPC框架，用于腿式机器人在避障中的共享自主。 quadruped legged robot legged locomotion
2 UnderwaterVLA: Dual-brain Vision-Language-Action architecture for Autonomous Underwater Navigation 提出UnderwaterVLA，用于水下自主导航，提升复杂环境下任务完成度。 MPC model predictive control vision-language-action
3 VLA-Reasoner: Empowering Vision-Language-Action Models with Reasoning via Online Monte Carlo Tree Search VLA-Reasoner：通过在线蒙特卡洛树搜索增强视觉-语言-动作模型的推理能力 manipulation imitation learning world model
4 MimicDreamer: Aligning Human and Robot Demonstrations for Scalable VLA Training MimicDreamer：对齐人类与机器人演示，实现可扩展的VLA模型训练 manipulation dreamer egocentric
5 Learning Multi-Skill Legged Locomotion Using Conditional Adversarial Motion Priors 提出基于条件对抗运动先验的多技能四足机器人运动学习框架 quadruped legged robot legged locomotion
6 VLBiMan: Vision-Language Anchored One-Shot Demonstration Enables Generalizable Bimanual Robotic Manipulation VLBiMan：基于视觉-语言锚定的单样本示教实现通用双臂机器人操作 manipulation bi-manual dual-arm
7 Developing Vision-Language-Action Model from Egocentric Videos 提出基于第一视角视频的视觉-语言-动作模型训练方法，无需人工标注。 manipulation teleoperation egocentric
8 Action-aware Dynamic Pruning for Efficient Vision-Language-Action Manipulation 提出动作感知动态剪枝ADP，提升视觉-语言-动作模型在机器人操作中的效率。 manipulation vision-language-action VLA
9 Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgetting 提出VLM2VLA，通过语言对齐解决VLM微调为VLA时的灾难性遗忘问题 teleoperation vision-language-action VLA
10 EgoDemoGen: Novel Egocentric Demonstration Generation Enables Viewpoint-Robust Manipulation EgoDemoGen：生成新颖的自我中心视角演示，实现视角鲁棒的机器人操作 manipulation imitation learning egocentric
11 Pixel Motion Diffusion is What We Need for Robot Control DAWN：基于像素运动扩散的机器人控制统一框架，实现语言条件下的机器人操作。 manipulation motion diffusion language conditioned ✅
12 Effect of Gait Design on Proprioceptive Sensing of Terrain Properties in a Quadrupedal Robot 步态设计影响四足机器人对地形属性的本体感受，提出适用于地形感知的步态策略。 quadruped legged robot locomotion
13 ARMimic: Learning Robotic Manipulation from Passive Human Demonstrations in Augmented Reality ARMimic：利用增强现实中被动的人类演示学习机器人操作 manipulation teleoperation imitation learning
14 From Watch to Imagine: Steering Long-horizon Manipulation via Human Demonstration and Future Envisionment Super-Mimic：结合人类演示和未来预测，实现长时程操作任务的零样本模仿学习 manipulation physically plausible multimodal
15 An Intention-driven Lane Change Framework Considering Heterogeneous Dynamic Cooperation in Mixed-traffic Environment 提出一种考虑异构动态合作的意图驱动型车道变换框架，用于混合交通环境。 model predictive control motion planning reinforcement learning
16 Towards Developing Standards and Guidelines for Robot Grasping and Manipulation Pipelines in the COMPARE Ecosystem 为COMPARE生态系统中的机器人抓取与操作流程开发标准与指南 manipulation motion planning
17 Multi-stage robust nonlinear model predictive control of a lower-limb exoskeleton robot 提出多阶段鲁棒非线性模型预测控制以解决下肢外骨骼机器人控制问题 MPC model predictive control
18 DemoGrasp: Universal Dexterous Grasping from a Single Demonstration DemoGrasp：基于单次演示的通用灵巧抓取方法 manipulation dexterous hand reinforcement learning
19 RoboView-Bias: Benchmarking Visual Bias in Embodied Agents for Robotic Manipulation 提出RoboView-Bias基准，用于量化具身智能体在机器人操作中的视觉偏差 manipulation
20 SAGE: Scene Graph-Aware Guidance and Execution for Long-Horizon Manipulation Tasks SAGE：基于场景图的长程操作任务引导与执行框架 manipulation
21 Robot Learning from Any Images RoLA：从任意图像生成交互式物理机器人环境，实现大规模机器人数据生成。 humanoid sim-to-real ✅
22 HELIOS: Hierarchical Exploration for Language-grounded Interaction in Open Scenes HELIOS：开放场景中基于语言交互的分层探索方法 manipulation mobile manipulation
23 Empart: Interactive Convex Decomposition for Converting Meshes to Parts Empart：交互式凸分解工具，实现网格到部件的区域定制简化，提升机器人仿真效率。 motion planning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
24 DHAGrasp: Synthesizing Affordance-Aware Dual-Hand Grasps with Text Instructions DHAGrasp：提出文本引导的双手抓取生成方法，实现语义感知的抓取合成 affordance affordance-aware ✅
25 Good Weights: Proactive, Adaptive Dead Reckoning Fusion for Continuous and Robust Visual SLAM 提出Good Weights算法以解决视觉SLAM在低纹理环境中的定位问题 visual SLAM scene understanding

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#题目一句话要点标签🔗⭐
26 Leveraging Large Language Models for Robot-Assisted Learning of Morphological Structures in Preschool Children with Language Vulnerabilities 利用大型语言模型辅助机器人为语言障碍幼儿提供形态结构学习 large language model
27 MINT-RVAE: Multi-Cues Intention Prediction of Human-Robot Interaction using Human Pose and Emotion Information from RGB-only Camera Data MINT-RVAE：利用RGB相机的人体姿态和情感信息进行人机交互意图预测 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
28 WoW: Towards a World omniscient World model Through Embodied Interaction WoW：通过具身交互构建世界全知世界模型，提升物理因果理解 world model
29 FlowDrive: moderated flow matching with data balancing for trajectory planning FlowDrive：结合数据平衡的适度流匹配轨迹规划，提升罕见场景性能。 flow matching

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
30 Multi-Robot Allocation for Information Gathering in Non-Uniform Spatiotemporal Environments 提出一种基于变差函数和信息度量的多机器人时空环境信息收集方法 spatiotemporal

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
31 An Ontology for Unified Modeling of Tasks, Actions, Environments, and Capabilities in Personal Service Robotics 提出OntoBOT本体，统一建模服务机器人任务、动作、环境与能力 spatial relationship

⬅️ 返回 cs.RO 首页 · 🏠 返回主页