cs.RO（2026-10-01）

📊 共 49 篇论文 | 🔗 10 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (37 🔗8) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱一：机器人控制 (Robot Control) (37 篇)

#题目一句话要点标签🔗⭐
1 InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation 提出InterEvolve以解决人形机器人在测试时任务演化问题 humanoid manipulation loco-manipulation
2 FlashDexRetarget: Accelerating Dexterous Manipulation Data Generation through Multi-Motion Retargeting 提出FlashDexRetarget以解决灵巧操控数据生成效率低下问题 motion tracking manipulation dexterous manipulation ✅
3 OrbitTAMP: Grounding Language Models for Task and Motion Planning in Spacecraft Rendezvous 提出OrbitTAMP以解决航天器会合任务与运动规划问题 trajectory optimization motion planning large language model
4 ReCo: Response-Consistent Locomotion with Policy-Aware MPC for Legged Manipulation 提出ReCo框架以解决连续腿部操作中的响应一致性问题 locomotion manipulation MPC
5 MASkillBlender: Decentralized Whole-Body Coordination for Multi-Humanoid Loco-Manipulation via Skill Blending 提出MASkillBlender以解决多智能体协调控制问题 humanoid whole-body control manipulation
6 PROMO: Preference-conditioned Multi-Objective Reinforcement Learning for Quadrupedal Robots 提出PROMO以解决四足机器人多目标强化学习问题 quadruped legged locomotion locomotion
7 NarrativeFlow: Flow-Based Vision-Language-Action Model Using Robot Velocity Fields 提出NarrativeFlow以解决语言条件下机器人操作的流动建模问题 manipulation flow matching vision-language-action ✅
8 eRLT: Efficient VLA Reinforcement Learning via Action-Relevant Token Routing 提出eRLT以解决VLA模型在强化学习中的样本效率问题 manipulation reinforcement learning vision-language-action
9 Is Success All You Need? Investigating the Impact of Input Perturbations on VLA Behaviour in Tabletop Manipulation Tasks 提出基准无关评估框架以提升VLA模型的行为鲁棒性 manipulation vision-language-action VLA
10 WBAG: A Whole-Body and Attached-Geometry Safety Framework for Vision-Language-Action Manipulation 提出WBAG框架以解决机器人视觉语言动作中的安全问题 manipulation vision-language-action VLA
11 Divide-and-Remember: Recursive Action-Relevant Memory for Long-Horizon VLA Policies 提出Divide-and-Remember以解决长时间依赖的VLA策略问题 manipulation imitation learning vision-language-action
12 Continue, Abort, or Fall: Viability-Aware Policy Selection (VAPS) for Safe Humanoid Acrobatics 提出VAPS以解决动态人形机器人安全控制问题 humanoid motion tracking sim-to-real
13 TOAST: Stochastic Robot Action Tokenization for Autoregressive Vision-Language-Action Models 提出TOAST以解决有限演示下的机器人策略学习问题 manipulation policy learning vision-language-action
14 HumanoidToolBench: Benchmarking Humanoid Tool Use from Selection to Mobile Execution 提出HumanoidToolBench以解决人形机器人工具使用评估问题 humanoid locomotion manipulation ✅
15 Experience-Based Feasibility-Aware Generative Adversarial Imitation from Observation under Embodiment Mismatch 提出EF-GAIfO以解决机器人模仿观察中的可行性问题 quadruped locomotion policy learning
16 EIDA: Execution-Interface Dynamics Adaptation for Real-to-Sim-to-Real Robot Navigation 提出EIDA以解决机器人导航中的真实与模拟转移问题 sim-to-real actuator dynamics Unitree
17 SkeleWAM: Skeleton World-Action Modeling for Efficient Robotic Manipulation 提出SkeleWAM以解决机器人操作中的世界动作建模问题 manipulation world action model world action models ✅
18 ColoACT: Multi-Cue Action Chunking for Smooth Autonomous Colon Navigation on a Self-Propelled Endoscopic Robot 提出ColoACT以解决自主结肠导航中的多重挑战 sim-to-real reinforcement learning deep reinforcement learning ✅
19 DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication 提出DuoMind以解决多机器人协调问题 manipulation vision-language-action VLA
20 Recova: Agent-Guided Failure Recovery for Autonomous Robotic Manipulation 提出Recova框架以解决自主机器人操作中的失败恢复问题 manipulation
21 Reconstruct, Practice, Go Real: Guided Self-Improvement for Embodied Agents 提出RPG框架以实现机器人自主技能提升 manipulation multimodal
22 ActiveWAM: Evidence-Aware Active Vision for World-Action Models 提出ActiveWAM以解决主动视觉中的证据感知问题 manipulation bi-manual
23 Are Frontier VLM Agents Ready to Be Robot Generalists? An Empirical Study with the Embodied Agent Arena 提出Embodied Agent Arena以评估前沿视觉语言模型的机器人通用性 manipulation affordance
24 Learning a Resolution-Consistent Jacobian Field for Bio-Inspired Rigid-Soft Finger 提出Jacobian流匹配方法以解决生物启发的刚柔指控制问题 trajectory optimization flow matching
25 Watch, Infer, Coordinate: Inferring Robot Partner Constraints for Zero-Shot Coordination 提出Watch, Infer, Coordinate以解决机器人协作中的物理约束推断问题 manipulation
26 Training-Free Diffusion Planning with Analytical Local Scores 提出无训练扩散规划方法以解决多机器人路径规划问题 motion planning
27 Query-Conditioned Articulation Estimation from a Single Image 提出QueryArt以解决单图像关节参数估计问题 manipulation ✅
28 quARtet Marker: A 3D-Printable Multi-Tag Fiducial for Robust Near-Frontal Pose Estimation 提出quARtet标记以解决透明物体定位问题 manipulation
29 Screw Attention: Rigid-Body Algebra Inside a Transformer 提出螺旋注意力机制以解决机器人操作中的几何变化问题 manipulation
30 SocialVLA: A Social Perception Gateway for Human-Reaction-Based Failure Detection and Recovery in VLA Manipulation 提出SocialVLA以解决VLA操作中的失败检测与恢复问题 manipulation Unitree vision-language-action
31 Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking 提出CoFiT方法以解决人形机器人安全全身跟踪问题 humanoid humanoid robot humanoid control
32 SoTa: Soft Tactile Skins for Dexterous Manipulation 提出SoTa软触觉皮肤以解决机器人灵巧操作中的触觉感知问题 manipulation dexterous manipulation cross-embodiment
33 Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation 提出ViGAR以解决长时间复合机器人操作问题 manipulation world model world models
34 RUL-Aware RRT*: Degradation-Balanced Motion Planning for Robotic Manipulators 提出RUL感知RRT*以解决机器人关节退化问题 motion planning
35 OpenRUA: Robot-Use Agents Are Zero-Shot Visuomotor Policies 提出OpenRUA以实现机器人控制的零-shot视觉运动策略 manipulation ✅
36 Query-Conditioned Articulation Estimation from a Single Image 提出QueryArt以解决单图像关节参数估计问题 manipulation ✅
37 Screw Attention: Rigid-Body Algebra Inside a Transformer 提出Screw Attention以解决刚体运动中的几何变化脆弱性问题 manipulation

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
38 Completion Aware Guidance for World Action Models 提出Completion Aware Guidance以解决WAMs任务完成问题 world model world models world action model
39 UniWAM: Unified World-Action Model 提出UniWAM以解决视觉-语言-动作模型的动态理解问题 flow matching egocentric spatiotemporal
40 Robot Learning on Discrete Surfaces: Theory and Applications 提出统一离散黎曼框架以解决机器人学习在多面体表面的问题 flow matching 3D reconstruction motion generation
41 TouchTherm: Building Multimodal Digital Twins of Objects for Tactile and Thermal Rendering 提出TouchTherm框架以构建多模态数字双胞胎解决触觉与热反馈问题 MAE multimodal
42 Kinematic MeanFlow: One-Step Action Generation Policy for Robotic Foundation Models 提出Kinematic MeanFlow以解决机器人基础模型中的高推理延迟问题 flow matching foundation model ✅
43 World Motion Models: Flexible Sequence Modeling of SE(3) Trajectories 提出世界运动模型以解决动态3D场景建模问题 policy learning flow matching cross-embodiment
44 Cross-entropy optimization with prioritized constraints 提出TierCEM以解决约束优先级冲突问题 world model world models
45 World-Calibrated Proposal-to-Action Flow for Vision-Language-Action Models 提出ProAct框架以解决视觉-语言-动作模型的生成连续性问题 flow matching vision-language-action VLA
46 TouchTherm: Building Multimodal Digital Twins of Objects for Tactile and Thermal Rendering 提出TouchTherm框架以构建多模态数字双胞胎，解决触觉与热反馈问题 MAE multimodal

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

#题目一句话要点标签🔗⭐
47 ChunkVLA-AM: Parallel Action Chunking for Vision-Language-Action Robot Control in Additive Manufacturing 提出ChunkVLA-AM以解决增材制造中的视觉-语言-动作机器人控制问题 vision-language-action VLA OpenVLA
48 UniTrackPLA: Unified Panorama-Language-Action Model for Instruction-Guided Navigation and Dynamic Person Tracking 提出UniTrackPLA以解决指令引导导航与动态人追踪问题 VLN ✅

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

#题目一句话要点标签🔗⭐
49 Real-time Event-camera Stereo Visual Odometry via Keytime Gaussian Process Regression 提出实时事件相机立体视觉里程计以解决估计状态过大的问题 visual odometry TAMP

⬅️ 返回 cs.RO 首页 · 🏠 返回主页