cs.RO（2026-05-15）

📊 共 24 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (18 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (1 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (18 篇)

#题目一句话要点标签🔗⭐
1 Learning Sim-Grounded Policies for Bimanual Rope Manipulation from Human Teleoperation Data 提出基于物理仿真的策略学习方法，提升双臂操作绳索解结任务的泛化性。 manipulation bi-manual teleoperation
2 DexJoCo: A Benchmark and Toolkit for Task-Oriented Dexterous Manipulation on MuJoCo DexJoCo：面向灵巧操作的MuJoCo基准测试与工具包 manipulation dexterous hand dexterous manipulation
3 Hierarchical and Holistic Open-Vocabulary Functional 3D Scene Graphs for Indoor Spaces 提出一种分层和整体的开放词汇功能性3D场景图方法，用于理解室内空间。 manipulation scene understanding open-vocabulary
4 Terrain Consistent Reference-Guided RL for Humanoid Navigation Autonomy 提出地形一致的参考引导强化学习方法，实现人形机器人自主导航 humanoid humanoid robot locomotion
5 Feedback World Model Enables Precise Guidance of Diffusion Policy 提出反馈世界模型，通过闭环反馈提升扩散策略的机器人控制精度 manipulation diffusion policy world model
6 Constrained MPC-Based Motion Planning for Morphing Quadrotors in Ultra-Narrow Passages under Limited Perception 提出一种基于约束MPC的运动规划方法，用于变形四旋翼在狭窄通道内的导航，解决感知受限问题。 MPC model predictive control motion planning ✅
7 NavRL++: A System-Level Framework for Improving Sim-to-Real Transfer in Reinforcement Learning-Based Robot Navigation NavRL++：用于提升基于强化学习的机器人导航中Sim-to-Real迁移的系统级框架 legged robot sim-to-real reinforcement learning
8 OHP-RL: Online Human Preference as Guidance in Reinforcement Learning for Robot Manipulation 提出OHP-RL框架，利用在线人类偏好指导机器人操作强化学习 manipulation reinforcement learning policy learning
9 GAP: Geometric Anchor Pre-training for Data-Efficient Visuomotor Learning of Manipulation Tasks GAP：用于操作任务数据高效视觉运动学习的几何锚点预训练 manipulation policy learning imitation learning
10 Adaptive Outer-Loop Control of Quadrotors via Reinforcement Learning 提出基于强化学习的四旋翼自适应外环控制，提升抗扰动能力。 sim-to-real domain randomization reinforcement learning
11 Learning Dynamic Pick-and-Place for a Legged Manipulator 提出一种基于强化学习的腿式机器人动态抓取放置框架，提升负载能力和工作空间。 quadruped whole-body control locomotion
12 SkiP: When to Skip and When to Refine for Efficient Robot Manipulation SkiP：针对机器人操作任务，提出动态跳跃与精细化策略提升效率 manipulation imitation learning behavior cloning ✅
13 Driving Through the Network: Performance and Workload Under Latency and Video Impairment 研究网络延迟和视频质量对远程驾驶性能与工作负荷的影响 manipulation teleoperation multimodal
14 FocalPolicy: Frequency-Optimized Chunking and Locally Anchored Flow Matching for Coherent Visuomotor Policy FocalPolicy：结合频率优化分块和局部锚定流匹配的连贯视觉运动策略 manipulation flow matching
15 FLASH: Efficient Visuomotor Policy via Sparse Sampling FLASH：基于稀疏采样的快速视觉运动策略学习 manipulation policy learning diffusion policy
16 KaRMA: A Kinematic Metric for Fine Manipulation Ability in Robotic Hands 提出KaRMA指标，评估机器人手在灵巧操作中的运动学能力 manipulation
17 Fast Expanding Safe Circular Regions for Efficient Local Path Planning 提出基于快速扩展安全圆形区域的局部路径规划方法，提升复杂环境导航效率。 model predictive control
18 Reactive Robot-Centric Safety for Autonomous Navigation in Constrained and Dynamic Environments 提出基于激光雷达感知的机器人中心安全控制架构，用于约束动态环境下的自主导航 quadruped

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
19 Health-Conditioned Vision-Language-Action Models for Malfunction-Aware Robot Control 提出健康状态感知的VLA模型，用于应对机器人功能退化下的控制任务 vision-language-action VLA ✅
20 Propagating Unsafe Actions in LLM Controlled Multi-Robot Collaboration via Single Robot Compromise 提出InfectBot攻击，通过单机器人入侵传播恶意行为至LLM控制的多机器人系统。 large language model ✅
21 Hybrid LLM-based Intelligent Framework for Robot Task Scheduling 提出基于混合LLM的智能框架，优化建筑机器人任务调度。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
22 WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation 提出WorldVLN，用于空中视觉语言导航的自回归世界行为模型。 reinforcement learning world model world models ✅

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
23 Beyond Collision Avoidance: Multi-Robot Yielding and Spatial Affordance in Emergency Evacuations 针对紧急疏散，提出考虑空间可供性和人类期望的多机器人避让策略 affordance

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

#题目一句话要点标签🔗⭐
24 A Topology-Aware Spatiotemporal Handover Framework for Continuous Multi-UAV Tracking 提出拓扑感知的时空切换框架，用于连续多无人机车辆跟踪。 spatiotemporal ✅

⬅️ 返回 cs.RO 首页 · 🏠 返回主页