cs.RO（2026-10-21）

📊 共 45 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (33 🔗4) 支柱九：具身大模型 (Embodied Foundation Models) (6) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (33 篇)

#题目一句话要点标签🔗⭐
1 SP-HLM: Sparse-Pose-Driven Motion Generation and Whole-Body Control for Humanoid Loco-Manipulation 提出SP-HLM框架以解决类人机器人运动生成与全身控制问题 humanoid humanoid robot whole-body control
2 HuMiT: Low-Latency Whole-Body Humanoid Teleoperation via Compact Reference Tracking 提出HuMiT以解决低延迟全身类人机器人遥操作问题 humanoid motion tracking manipulation ✅
3 Watch Your Step: Anticipated Perception for Humanoid Locomotion 提出轻量化预测模型以提升人形机器人步态控制 humanoid humanoid locomotion locomotion
4 AtomicRL: Atomic Reinforcement Learning for Long-Horizon Robotic Manipulation 提出AtomicRL以解决长时间机器人操作中的信用分配问题 manipulation sim2real real2sim
5 SMR: Learning streaming motion retargeting for humanoids from human semantic 提出SMR以解决人类运动向类人机器人重定向问题 humanoid whole-body control motion retargeting
6 Agentic World Action Models 提出Agentic世界动作模型以解决技能重用不足问题 manipulation bi-manual bimanual manipulation
7 Test-Time Motion Steering for Perceptive Humanoid Fall Recovery 提出RISE以解决人形机器人跌倒恢复问题 humanoid fall recovery Unitree ✅
8 A2I-Calib++: A Kinematics-Driven Anti-noise Active Calibration Framework for Whole-Body Multi-IMU Systems on Legged Robots 提出A2I-Calib++以解决腿部机器人多IMU系统的噪声校准问题 quadruped legged robot locomotion
9 Do Earth-Trained Humanoid Controllers Walk on the Moon? Gravity, Terrain, and Soil Transfer 提出一种新方法以解决人形机器人在月球行走的适应性问题 humanoid humanoid control locomotion
10 PHD-Hand: A Printable Humanoid Dexterous Hand with Linkage-Driven 3-DOA Digits 提出一种可打印的人形灵巧手以解决传统手部设计的局限性 humanoid humanoid robot manipulation
11 SimPT: An Automated Simulation Framework for Intervention-Based VLA Post-Training 提出SimPT框架以自动化干预式VLA后训练 manipulation DPO vision-language-action
12 CW-WAM: A Structured Contact-Wrench World Action Model for Force-Sensitive Manipulation 提出CW-WAM以解决力敏感操作中的接触预测问题 manipulation world action model world action models
13 DONIC: Generalist Dexterous Bimanual Tracking Control from Human Demonstrations 提出DONIC以解决双手灵巧追踪控制问题 manipulation bi-manual bimanual manipulation
14 From VLM Planners to Robotic Agents: Self-Evolving Harnesses for History-Dependent Manipulation 提出自演化机器人代理以解决历史依赖的操控问题 manipulation vision-language-action VLA
15 Training in Deployment: Human-in-the-loop Online Post-Training for Robot Manipulation 提出TIDE以解决机器人操作中的在线后训练问题 manipulation bi-manual bimanual manipulation
16 Relational Geometry Forcing for World Action Models 提出关系几何强制方法以解决世界动作模型的几何约束问题 manipulation world action model world action models
17 Integrating Flow Matching and Model Predictive Control for Safe Robot Navigation 提出流匹配与模型预测控制结合以解决安全机器人导航问题 model predictive control flow matching multimodal
18 DepState-JEPA: Learning Depth-Grounded State Representation for Visual Perception and Manipulation 提出DepState-JEPA以解决视觉感知与操控中的状态表示问题 manipulation JEPA Joint-Embedding Predictive Architecture
19 End-to-End Reinforcement Learning for Quadrupedal Jumping in Lunar Gravity 提出多阶段强化学习方法以实现月球重力下的四足跳跃 quadruped legged robot reinforcement learning
20 Distance Field Driven Reinforcement Learning for Reactive and Robust Collision Avoidance 提出基于距离场的强化学习以解决碰撞避免问题 sim-to-real reinforcement learning zero-shot transfer
21 Asymmetric On-Policy Distillation Learns In-Context Exploration 提出ASTEROID框架以解决部分可观测性下的探索问题 manipulation reinforcement learning distillation
22 CausGround: Causal Interaction Grounding for Adaptive Embodied Planning 提出CausGround以解决多模态大语言模型的物理交互规划问题 manipulation large language model multimodal
23 Linear Twisting Sidewinding Gaits for Twistable Snake Robot Motion Planning on Complex Terrains 提出线性扭转侧行步态以解决蛇形机器人运动规划问题 locomotion motion planning
24 Plan First, Act Later: Streaming Reasoning for Executable Memory in Robot Control 提出PAL框架以解决长时间机器人操控中的记忆问题 manipulation vision-language-action VLA ✅
25 RangeNav: Task-Aligned LiDAR Range-Image Pretraining for Robot Navigation 提出RangeNav以解决稀疏奖励下的导航任务 quadruped Unitree reinforcement learning
26 From Reasoning to Action: Benchmarking Frontier AI Models on LIBERO-10 Robot Manipulation Tasks 提出统一框架评估前沿AI模型在机器人操作任务中的表现 manipulation
27 Reconstruction-Augmented Visuomotor Pre-Training of 3D-Grounded Bimanual Dexterity from RGB-Only Human Videos 提出SpatialDex以解决RGB视频中三维双手灵巧性学习问题 bi-manual
28 Preserving Operator Intent Under Actuation Constraints: Teleoperation of Manipulators with Coupled Transmissions 提出基于执行约束的遥操作方法以解决机器人控制问题 teleoperation
29 PRISM: Learning Realistic Depth via Physics-Grounded Noise Disentanglement with Semantic-Geometric Collaboration 提出PRISM以解决真实深度合成中的噪声问题 sim-to-real foundation model ✅
30 Object- and Task-Aware Active Perception for Safe Navigation of Mobile Manipulators in Semi-Static Environments 提出主动感知框架以解决移动操纵器在半静态环境中的安全导航问题 model predictive control
31 Toward Recursive Self-Improvement in Physical AI 提出递归自我改进机制以提升物理AI的学习能力 manipulation
32 Make it Walk: Computational Design of LEGO Walkers 提出一种计算设计方法以实现多种步态的LEGO步行器 locomotion
33 S-Rolling: Steerable Rolling Gaits for Twistable Snake Robots in Low-Clearance Environments 提出S-Rolling以解决低净空环境下蛇形机器人灵活性问题 locomotion

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#题目一句话要点标签🔗⭐
34 Post-hoc Epistemic Uncertainty for Vision-Language-Action Policies 提出后处理内在不确定性估计以优化视觉-语言-行动策略 vision-language-action VLA
35 NavBind: Diagnosing Instruction-Action Binding at Route Forks in Navigation VLAs 提出NavBind以解决导航VLA模型指令与动作绑定问题 vision-language-action VLA
36 Multimodal Embodied Agents: A Unified Capability-centric Perspective from Computer-Use to Robot-Use 提出PAPAV框架以解决多模态代理在机器人应用中的决策问题 multimodal
37 TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social Environments 提出TARSphere框架以解决机器人在动态环境中的适应性问题 large language model multimodal
38 FlowVLA: Learning Flow-Aware Physical Completion for Quantitative Robotic Pouring 提出FlowVLA以解决机器人倒液过程中的质量估计问题 vision-language-action VLA
39 OMAR: Open-World Multi-Attribute Reasoning for Scene-Conditioned Shelf Organization 提出开放世界多属性推理以解决场景条件下的货架组织问题 foundation model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
40 How Representational Priors Shape Generalization in World Action Models 提出表征先验以提升世界行动模型的泛化能力 world model world models world action model
41 PhysiCalWorld: In-Context Physics Calibration for World Models 提出PhysiCalWorld以解决世界模型物理属性校准问题 world model world models
42 Mixture of Flow Matching and Particle Filter for Proprioception-Based Online Contact Estimation 提出ContactFlowFilter以解决机器人接触估计问题 flow matching
43 State-Dependent Objective Coordination for End-to-End Balance and Navigation 提出状态依赖目标协调方法以解决机器人平衡与导航问题 reinforcement learning PPO

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

#题目一句话要点标签🔗⭐
44 AirPlayground: Benchmarking Aerial Navigation across Diverse Reconstructed Environments 提出AirPlayground以解决无人机导航评估与训练问题 3D gaussian splatting 3DGS gaussian splatting

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

#题目一句话要点标签🔗⭐
45 From Acceptance and Avoidance to Affective Viscosity: Designing Social Agents Intermediating in Human Relationships 提出情感粘性模型以改善社交代理在人际关系中的作用 spatiotemporal

⬅️ 返回 cs.RO 首页 · 🏠 返回主页