cs.RO（2026-05-29）

📊 共 27 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (19 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱一：机器人控制 (Robot Control) (19 篇)

#题目一句话要点标签🔗⭐
1 DeMaVLA: A Vision-Language-Action Foundation Model for Generalizable Deformable Manipulation DeMaVLA：用于可变形物体操作的通用视觉-语言-动作基础模型 manipulation dual-arm flow matching
2 HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model 提出HARP框架，解决人-机器人视觉表征对齐问题，提升VLA模型性能 manipulation policy learning representation learning
3 Feat2Go: Visual Feature-Grounded Value Estimation for Embodied Reinforcement Learning Feat2Go：基于视觉特征的价值估计，提升具身强化学习性能 manipulation bi-manual bimanual manipulation
4 Learning Terrain-Aware Whole-Body Control for Perceptive Legged Loco-Manipulation 提出TA-WBC，一种地形感知的腿足机器人全身控制框架，用于复杂地形下的移动操作任务。 whole-body control WBC manipulation
5 A study on a Real-Time VR-Based Teleoperation Framework for Manipulator in Dynamic Environment 提出基于VR的实时遥操作框架，用于动态环境中机械臂的控制与避障 manipulation trajectory optimization teleoperation
6 TARIC: Memory-Augmented Traversability-Aware Outdoor VLN under Interrupted Semantic Cues 提出TARIC，解决户外VLN中语义线索中断下的可通行性导航问题 quadruped traversability VLN
7 RDGen: Demonstration Generation for High-Quality Robot Learning via Reinforcement Learning RDGen：通过强化学习生成高质量机器人学习演示数据 sim-to-real teleoperation reinforcement learning
8 LiftNav: Path Planning via Semantic Lifting in TSDF-Guided Gaussian Splatting LiftNav：基于TSDF引导高斯溅射的语义提升路径规划 trajectory optimization gaussian splatting splatting
9 Wall-OSS-0.5 Technical Report Wall-OSS-0.5：开源4B规模VLA模型，实现机器人零样本可执行行为 manipulation policy learning flow matching
10 IDOL: Inverse-Dynamics-Guided Future Prediction for End-to-End Autonomous Driving 提出IDOL框架以解决未来预测与轨迹优化的耦合问题 trajectory optimization world model world models
11 GSAM: A Generalizable and Safe Robotic Framework for Articulated Object Manipulation GSAM：一种通用且安全的铰链物体操作机器人框架 manipulation motion planning policy learning
12 High-Load-Density Electro-Permanent Magnetic Foot with Controllable Adhesion for Quadruped Wall-Climbing Robots 提出高负载密度电永磁吸盘，实现四足机器人可靠的墙壁攀爬 quadruped locomotion Unitree
13 SSR: Scaling Surefooted and Symmetric Humanoid Traversal to the Open World SSR：面向开放世界，提升双足人形机器人稳健性和对称性的步态规划 humanoid locomotion egocentric
14 Surface Constraint Policy for Learning Surface-Constrained and Dynamically Feasible Robot Skills 提出表面约束策略，解决机器人复杂表面操作中几何约束和动态可行性问题 manipulation dexterous manipulation imitation learning
15 Object-Informed Model Predictive Path Integral Control for Non-Prehensile Robot Manipulation 提出对象引导的MPPI控制，解决非抓取操作中的长程规划问题 manipulation
16 Batched Differentiable Rigid Body Dynamics in PyTorch for GPU-Accelerated Robot Learning BARD：用于GPU加速机器人学习的PyTorch批量可微刚体动力学库 quadruped reinforcement learning AMP ✅
17 Haptic Sorter: A Unified Planning Framework for Online Shape Estimation and Real-Time Pose Inference 提出Haptic Sorter，用于在线形状估计和实时姿态推断的统一规划框架 manipulation motion planning
18 Two Degree-of-Freedom Vibratory Transport in a Grasp 提出基于非对称振动的二自由度夹持内操作方法，实现零件的精确操控。 manipulation in-hand manipulation
19 Learning Controlled Separation of Small Objects Between Two Fingers with a Tactile Skin 提出一种基于触觉皮肤的强化学习方法，用于机器人灵巧手控制分离微小物体 sim-to-real reinforcement learning

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

#题目一句话要点标签🔗⭐
20 Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring 提出Hide-and-Seek框架，用于视觉-语言-动作模型（VLA）的运行时故障检测。 vision-language-action VLA OpenVLA
21 Primitive Subspaces Mediate Few-Shot Transfer in VLAs 提出原始子空间以解决视觉-语言-动作任务的少样本迁移问题 vision-language-action VLA OpenVLA
22 Can Aerial VLA Models Cooperate? Evaluating Closed-Loop Air-Ground Coordination with CARLA-Air CARLA-Air：评估空中VLA模型在空地协同中的闭环性能 vision-language-action VLA ✅

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

#题目一句话要点标签🔗⭐
23 AR Forcing: Towards Long-Horizon Robot Navigation World Model AR Forcing：面向长时程机器人导航的世界模型自回归训练 world model world models
24 Enhancing Human-Likeness in Reinforcement Learning Agents via Hierarchical Macro Action Quantization 提出层次化宏动作量化方法以增强强化学习代理的人类相似性 reinforcement learning SAC IQL

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

#题目一句话要点标签🔗⭐
25 Seeing Fast and Slow: Bimodal 3D Scene Graphs for Open-set Tasks BiMoSG：用于开放集任务的双模态3D场景图生成方法 open-vocabulary open vocabulary
26 On-Device Robotic Planning: Eliminating Inference Redundancy for Efficient Decision-Making 提出REIS框架，消除推理冗余，加速端侧机器人语义规划决策 affordance

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

#题目一句话要点标签🔗⭐
27 Don't Fool Me Twice: Adapting to Adversity in the Wild with Experience-Driven Reasoning 提出基于经验推理的自适应框架，提升机器人在未知环境中的鲁棒性 cross-embodiment

⬅️ 返回 cs.RO 首页 · 🏠 返回主页