cs.RO（2026-11-02）

📊 共 45 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (35 🔗7) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱三：空间感知与语义 (Perception & Semantics) (3 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱一：机器人控制 (Robot Control) (35 篇)

#题目一句话要点标签🔗⭐
1 MAPO: Mixture of Experts with Alternating Policy Optimization for Universal Humanoid Motion Tracking 提出MAPO以解决人形机器人运动跟踪中的政策更新问题 humanoid humanoid robot motion tracking ✅
2 Pixel: A Co-Designed Full-Stack Robot System for Learning General-Purpose Mobile Manipulation 提出Pixel以解决通用移动操作中的系统集成问题 humanoid whole-body control manipulation
3 ReGATE: Recurrent Gated Visual Fusion for Reactive Humanoid-Object Interaction 提出ReGATE以解决人形机器人在视觉中断下的物体交互问题 humanoid whole-body control locomotion
4 TAMPAgent: Agentic Belief-Space Planning for Mobile Manipulation 提出TAMPAgent以解决开放世界移动操控中的不确定性问题 manipulation mobile manipulation motion planning
5 ReContact: Reconstruct and Retarget Human Contacts from Video for Contact-Rich Humanoid Manipulation 提出ReContact以解决人形机器人接触重建与重定向问题 humanoid humanoid robot manipulation
6 Beyond Policy Patches: Distilling RL Specialists into Vision-Language-Action Policies 提出REPAIR框架以提升视觉-语言-动作策略在精确操作中的表现 manipulation reinforcement learning flow matching
7 Pixel-UMI: Learning Environment-Aware Whole-Body Manipulation from Sensorized Human Demonstrations 提出Pixel-UMI以解决传感器化人类示范中的全身操控问题 manipulation mobile manipulation whole-body manipulation
8 ForceLDM: A Force-aware Latent Dynamics Model for Contact-Rich Manipulation 提出ForceLDM以解决接触丰富操控中的动态建模问题 manipulation policy learning latent dynamics ✅
9 SomaAlign: Reliability-Aware Multimodal Representation Learning for Soft Robotic Grasping 提出SomaAlign以解决软机器人抓取中的多模态感知挑战 manipulation representation learning multimodal
10 DexArena: A Unified Benchmark for Long-Horizon Dexterous Manipulation 提出DexArena以解决长时间灵巧操作基准评估问题 manipulation dexterous manipulation bi-manual
11 FineCraft: Enhancing VLA Models for Fine-Grained Manipulation 提出FineCraft以解决细粒度操作中的视觉与指令对接问题 manipulation VLA ✅
12 VLALagBench: Understanding When and Why VLA Inference Latency Matters 提出VLALagBench以研究VLA推理延迟对任务成功率的影响 manipulation vision-language-action VLA
13 KlaskTron-IL: A Contact-Rich Adversarial Imitation Learning Benchmark with a Validated Sim-to-Real Bridge 提出KlaskTron-IL以解决接触丰富环境下模仿学习的挑战 sim-to-real reinforcement learning PPO
14 RoboColosseum: Evaluating Physical Intelligence at Scale In-The-Wild 提出RoboColosseum以解决机器人政策评估的可重复性问题 humanoid bi-manual cross-embodiment
15 HeLM: Hierarchical Explicit Language Memory for Non-Markovian Manipulation 提出HeLM以解决非马尔可夫环境中的工作记忆问题 manipulation vision-language-action VLA
16 BeyondContact: A Reinforcement Learning Framework for Airflow-Based Non-Contact Manipulation 提出基于气流的非接触操控框架以解决接触操控限制问题 manipulation reinforcement learning ✅
17 Visuo-tactile Diffusion Policy for Robotic Manipulation with Auxiliary Tactile Dynamics Prediction 提出视触扩散策略以解决长时间机器人组装问题 manipulation diffusion policy
18 When Robots Outweigh Objects: How Inertia Mismatch Shapes Robot Policy Learning 提出方法以解决机器人与物体间的惯性不匹配问题 manipulation reinforcement learning policy learning
19 Learning Purposeful Co-designs of Bio-inspired Quadrupedal Robots 提出生物启发的四足机器人共设计方法以提升适应性 quadruped legged robot reinforcement learning
20 Exploring Proximity and Tactile Sensing for Robotic Manipulation 提出ToFTouch以解决机器人抓取前触觉感知问题 manipulation multimodal
21 Safe-Ballbot: Tilt-Consistent MPC with Control Barrier Functions for Ballbot Navigation 提出安全球机器人导航方案以解决动态障碍物避让问题 MPC model predictive control
22 Energy-Based Policy Orchestration 提出能量基础的策略编排以解决动态任务中的控制问题 humanoid manipulation loco-manipulation ✅
23 W2A: Distilling World Priors into Action Models for Real-Time Robotic Control 提出W2A框架以解决实时机器人控制中的计算延迟问题 manipulation world action model world action models
24 PanoTac: A Dense Three-Axis Tactile Sensor for Robust Touch-Centric Manipulation 提出PanoTac以解决触觉反馈不足的问题 manipulation
25 VerifLM: Physically Verified Image-Edit-Guided Goal Pose Generation for Robotic Manipulation 提出VerifLM以解决机器人操作中的目标姿态生成问题 manipulation
26 TRAILS: Trajectory Generation for Automated Indoor Locomotion Studies 提出TRAILS框架以自动生成室内导航场景 locomotion ✅
27 PointPilot: Unified Robotic Manipulation via Object-Centric 3D Point Flow Prediction 提出PointPilot以解决机器人操作中的泛化问题 manipulation
28 A Study on the Effects of Finger Movements on a Humanoid Hugging Robot 研究手指动作对类人拥抱机器人用户影响 humanoid
29 KlaskTron: An Open-Source Platform for Physical Adversarial Multi-Agent RL 提出KlaskTron平台以解决物理对抗多智能体强化学习的基准缺乏问题 manipulation sim-to-real reinforcement learning ✅
30 Mediating Human-Swarm Interactions Through a Robot Helper: A Shepherding-Inspired Approach 通过机器人助手调解人群-群体交互以解决群体控制问题 teleoperation HSI
31 Clinician-AI Collaboration in Endovascular Neurointerventions: In-Vitro User Study of Copilot Assistance Modes for Robotic Navigation Across the Autonomy Spectrum 提出Copilot辅助模式以提升机器人神经介入导航的临床协作 sim-to-real shared control
32 Closing the Dynamics Gap through Self-Directed Robotic Interaction 提出CUE以解决动态差距问题 sim2real real2sim
33 SNAP: Semantic Navigation Among Pedestrians 提出一种新型导航策略以解决拥挤环境中的行人避让问题 legged robot reinforcement learning
34 Sequential Recovery Updates for Learned Controllers under Successful-Trajectory Preservation 提出顺序恢复更新以提升学习控制器的成功轨迹保留能力 humanoid
35 Neuro Inspired Robotics: What Robot Learning Can Borrow from the Sensorimotor Brain 提出神经启发式机器人设计以解决日常操作中的学习不足问题 manipulation

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
36 Seeing Through the Illusion: Test-Time Scaling for World Action Models 提出WISE以解决世界动作模型的决策有效性问题 world action model world action models
37 Toward Uncertainty-Aware, Physics-Grounded Human World Models 提出不确定性感知的物理基础人类世界模型以提升机器人交互能力 world model world models
38 Right Ranking, Wrong Scale: Reproducing the Robotic World Model's Dynamics Model and Testing the Uncertainty of Its Follow-up 重现机器人世界模型的动态模型并测试不确定性估计 world model world models
39 An intelligent satellite must do nothing well 提出基于轨道动态的智能卫星自主检查方法 reinforcement learning world model world models
40 To Coordinate or Not to Coordinate? Learning to Swim with Eight Soft Arms 提出独立学习八臂游泳的控制策略以解决协调问题 reinforcement learning PPO

🔬 支柱三：空间感知与语义 (Perception & Semantics) (3 篇)

#题目一句话要点标签🔗⭐
41 MS-DOTT: A Multi-Modal Dataset for Off-Road Traversability Estimation 提出MS-DOTT以解决越野环境中的可通行性估计问题 traversability
42 AnchorHEC: Targetless Hand-Eye Calibration of Wrist and Fixed Cameras 提出AnchorHEC以解决无目标手眼标定问题 3D reconstruction foundation model ✅
43 SyncHeC: Markerless Spatio-Temporal Hand-Eye Calibration via Differentiable Rendering and Continuous-Time Joint Optimization 提出SyncHeC以解决无标记手眼标定问题 metric depth

🔬 支柱九：具身大模型 (Embodied Foundation Models) (1 篇)

#题目一句话要点标签🔗⭐
44 Source-Resolved Collective Cognition in Robotics: Mathematical Foundations for Multimodal and Collective Estimation, and Their Reach across the Cognition Loop 提出广义奇异值分解以解决多模态集体认知问题 multimodal

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

#题目一句话要点标签🔗⭐
45 A Tightly Coupled Magnetic-Inertial SLAM System for Field Robots 提出紧耦合磁惯性SLAM系统以解决恶劣环境下定位问题 motion estimation

⬅️ 返回 cs.RO 首页 · 🏠 返回主页