cs.RO（2026-10-16）

📊 共 39 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (27 🔗5) 支柱九：具身大模型 (Embodied Foundation Models) (5 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱三：空间感知与语义 (Perception & Semantics) (2)

🔬 支柱一：机器人控制 (Robot Control) (27 篇)

#题目一句话要点标签🔗⭐
1 BeJEPA: A Goal-Conditioned Humanoid Behavior World Model via Unsupervised RL 提出BeJEPA以解决类人机器人控制中的目标定位问题 humanoid humanoid control motion tracking ✅
2 CoEvolve: Self-Evolving Code Policies from Skills to Strategies for Humanoid Robots 提出CoEvolve框架以解决多机器人协作中的协调适应问题 humanoid humanoid robot manipulation
3 Reinforcement Learning Based Multi-Terrain Adaptive Whole-Body Control for a Quadruped Manipulator 提出基于强化学习的全身控制框架以解决四足机器人在复杂地形中的操控问题 quadruped legged robot whole-body control
4 MambaAct: Efficient Mamba Action Experts for Flow-Matching Vision-Language-Action Models 提出MambaAct以提高流匹配视觉-语言-动作模型的效率 manipulation flow matching Mamba ✅
5 GRASP: Global-Local Representation Alignment and Spatial-Physical Grounding for Bimanual Robotic Manipulation 提出GRASP框架以解决双手机器人操控中的任务成功率问题 manipulation bi-manual bimanual manipulation
6 Foundation-Model-Driven 3D Scene Graphs for Mobile Manipulation 提出基础模型驱动的3D场景图以解决移动操作中的空间记忆问题 manipulation mobile manipulation foundation model
7 Embodiment-Aware Planning for Heterogeneous Bimanual Manipulation 提出异构双手操作的具身感知规划以解决复杂操控问题 manipulation bi-manual dual-arm
8 Look Before You Act: Bridging Active Perception and Vision-Language-Action Models via Observation-Aligned Data Adaptation 提出ActiveAlign以解决机器人操作中的观察状态分布敏感性问题 manipulation vision-language-action VLA
9 ORD-WM: Learning Object-Centric Rigid Dynamics for 3D World Modeling in Robotic Manipulation 提出ORD-WM以解决机器人操作中的物体中心刚性动力学建模问题 manipulation model predictive control world model
10 No Eyes, No Problem: What Does Touch Contribute to Blind Baoding Ball Manipulation? 提出无视力的触觉反馈机制以解决盲操控问题 manipulation dexterous hand dexterous manipulation
11 Ego2Contact: Predicting Touch from Egocentric Human Video for Contact-Rich Robot Manipulation 提出Ego2Contact以解决机器人触觉感知不足问题 manipulation bi-manual reinforcement learning
12 Enhancing Visual Task-Redirection Attacks on Vision-Language-Action Models via Closed-Loop Optimization 提出TRACO框架以解决视觉语言行动模型的攻击问题 manipulation vision-language-action VLA
13 VLA-Fabric: Learning Interaction for Composable Multi-Arm VLA Agents 提出VLA-Fabric以解决多臂机器人协调问题 manipulation vision-language-action VLA
14 AWAM: Active Trajectory-Conditioned View Imagination for Robust Robot Manipulation 提出AWAM以解决开放世界机器人操作中的感知缺口问题 manipulation world model world models
15 Blind Precise Assembly: Proprioceptive Sim-to-Real Peg-in-Hole for Diverse Parts 提出盲目精确装配方法以解决复杂部件组装问题 legged robot humanoid sim-to-real
16 MOSS: Multimodal Origami Swamp Robot System 提出MOSS以解决湿地监测中的多样化地形适应问题 locomotion multimodal
17 LAW: Unifying Language, Action, and World Modeling for Agentic Physical Intelligence 提出LAW框架以统一语言、动作与世界建模，提升物理智能表现 manipulation world model world models
18 Foundation Models for Autonomous Space Robotics: Architectures, Physical Adaptation, and Validation 提出基础模型以解决自主空间机器人任务执行问题 manipulation foundation model
19 Pedicel-Targeted Strawberry Harvesting via Imitation Learning on a Low-Cost Teleoperated Robotic Arm 提出基于模仿学习的草莓 pedicel 目标采摘方法以解决精准抓取问题 manipulation motion planning teleoperation ✅
20 Robotic Spatiotemporal Editing: A Unified Interface for Controllable Action Generation via Masked Action Reconstruction 提出MAP接口以解决机器人行为控制与编辑问题 manipulation spatiotemporal
21 Sample-Efficient Reinforcement Learning for Policy Improvement via Flow-Reversed Action Anchors 提出AnchorQ以提高强化学习样本效率解决机器人策略改进问题 manipulation reinforcement learning flow matching
22 Shape-Aware Neural Time Fields for Rigid-Body Motion and Manipulation Planning 提出形状感知神经时间场以解决刚体运动规划问题 manipulation motion planning
23 DefFlow: Goal-Conditioned Deformable Object Manipulation using 3D Point Flows 提出DefFlow以解决变形物体目标导向操控问题 manipulation
24 InstAgent: A Recursively Self-Improving Agent Operating System for Heterogeneous Robot Collaboration 提出InstAgent以解决异构机器人协作中的执行时间修正和上下文记忆问题 manipulation multimodal
25 Not All Actions Contribute Equally: Learning Precise Progress Estimation from Imprecise Demonstrations 提出PRIME以解决机器人操作中的进度估计问题 manipulation reinforcement learning
26 Learning Robot Metareasoning: Adaptive Computation Budgets for Real-Time Control 提出自适应计算预算以解决实时控制中的规划与反应问题 manipulation reinforcement learning ✅
27 HexaShell: A Thin Harness for Exploring the Limits of VLM Agents as Robot Coding Agents 提出HexaShell以探索VLM代理作为机器人编码代理的极限 manipulation ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#题目一句话要点标签🔗⭐
28 Unifying Embodied Foundation Models via a Model-to-Policy Formulation 提出统一的模型到策略框架以整合多模态机器人学习 vision-language-action VLA foundation model
29 Beyond Success Rate: Interventional Evaluation of Memory in Vision-Language-Action Systems 提出干预性记忆评估协议以解决视觉-语言-行动系统中的记忆效用问题 vision-language-action VLA
30 From Sliding Windows to Self-evolving Lifelong Memory: A Dual-perspective Survey of Memory Paradigms for Long-Horizon Robotics VLA Models 提出系统化记忆机制以解决长时间机器人任务中的经验积累问题 vision-language-action VLA
31 NestedVLA: Learning to Consolidate and Generate Skills for Vision-Language-Action Models 提出NestedVLA以解决视觉-语言-动作模型技能整合问题 vision-language-action VLA ✅
32 PASS-VLN: Passability-Aware Local-Command Vision-and-Language Navigation 提出PASS-VLN以解决机器人局部导航可执行性问题 VLN

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

#题目一句话要点标签🔗⭐
33 SimpleVLN: Towards Simple and Frontier Vision-Language Navigation 提出SimpleVLN以解决视觉语言导航中的架构限制问题 reinforcement learning VLN large language model
34 GR-VIP: Geometric representation conditioning for precise visuomotor policy 提出GR-VIP以解决移动机器人精确控制问题 VIP zero-shot transfer
35 Beyond Action Prediction: Learning Search-Free Planning in Latent Space 提出HIJP框架以解决搜索优化规划的高成本问题 world model world models JEPA
36 Reason-guided Human Reward Modeling for Preference-based Reinforcement Learning 提出基于理由引导的人类奖励建模以解决偏好强化学习问题 reinforcement learning
37 Hardware-as-Regularization: Enhancing Data Consistency in Imitation Learning via an Adjustable Passive Compliant Leader Arm 提出硬件正则化方法以提升模仿学习中的数据一致性 imitation learning

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

#题目一句话要点标签🔗⭐
38 FLoTS-SLAM: Optical Flow-Derived Uncertainty for Monocular Dynamic SLAM with Two-Stage Optimization 提出FLoTS-SLAM以解决动态环境下单目SLAM问题 3D reconstruction optical flow
39 F-LIVO: A Feature-Level Geometric Measurement Framework for LiDAR-Inertial-Visual Odometry 提出F-LIVO框架以解决LiDAR-惯性-视觉里程计的资源消耗问题 visual odometry

⬅️ 返回 cs.RO 首页 · 🏠 返回主页