cs.RO（2026-09-22）

📊 共 55 篇论文 | 🔗 11 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (43 🔗10) 支柱三：空间感知与语义 (Perception & Semantics) (4) 支柱九：具身大模型 (Embodied Foundation Models) (4) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱四：生成式动作 (Generative Motion) (1 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (43 篇)

#题目一句话要点标签🔗⭐
1 PLAT: Sparse Timed Keyframe Motion Tracking for Humanoid Control via Privileged Latent Transition Learning 提出PLAT以解决稀疏时序关键帧运动跟踪问题 humanoid humanoid robot humanoid control
2 VisForce: Visual Grounding of Current and Desired Forces for Goal-Conditioned Dexterous Manipulation 提出VisForce以解决力与视觉位置对应问题 manipulation dexterous hand dexterous manipulation
3 MATE: Multi-Agent Virtual Teleoperation Platform for Humanoid Collaboration Data Collection 提出MATE平台以解决人形机器人协作数据收集问题 humanoid humanoid robot manipulation ✅
4 MedVLA: A Hierarchical Vision-Language-Action Framework for Closed-Loop Precision Medical Robot Manipulation 提出MedVLA框架以解决精密医疗机器人操作中的决策问题 manipulation vision-language-action VLA
5 Sample, Simulate, Select: Physics-in-the-Loop Text-to-Motion for Humanoids Without Training 提出S$^3$方法以解决无训练的人形机器人动态控制问题 humanoid Unitree text-to-motion
6 The Cartesian Hand: In-Hand Manipulation with All-Linear Fingers 提出Cartesian Hand以解决灵巧抓取与相对操作的复杂性问题 humanoid manipulation dexterous hand
7 SafeLoop: Risk-Aware Rollback for Vision-Language-Action Manipulation 提出SafeLoop以解决视觉-语言-动作操作中的风险问题 manipulation vision-language-action VLA ✅
8 Teaching Reinforcement Learning and Humanoid Robotics to High-School Students: An Expert-Validated Curriculum Design on a Low-Cost Open Platform 提出一套课程框架以解决高中生机器人学习的挑战 humanoid humanoid robot reinforcement learning
9 RouteRLT: Learning When and Which RL Specialist Should Control a Vision-Language-Action Policy 提出RouteRLT以解决视觉-语言-动作模型在精确控制中的挑战 manipulation reinforcement learning behavior cloning
10 SG-CPG: Severity-Gated Central Pattern Generators for Adaptive Quadruped Locomotion under Continuous Actuator Degradation 提出SG-CPG以解决四足机器人在持续执行器退化下的适应性运动问题 quadruped locomotion locomotion policy
11 HABILIS Brain 0: Geometry-Change Supervision for Vision-Language-Action and Residual Flow Recovery 提出GC-VLA以解决几何监督不足的问题 manipulation egocentric vision-language-action
12 Imperfection for Precision: Upcycling Imperfect Data for High-Precision Robotic Manipulation 提出$4P$方法以利用不完美数据提升机器人高精度操作 manipulation teleoperation flow matching ✅
13 MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence 提出MachEmbodied-U0以解决通用机器人控制中的理解与生成问题 manipulation flow matching affordance
14 TriWorldBench: A Tri-View Consistency Perspective on Embodied World Models 提出TriWorldBench以解决机器人多视角一致性评估问题 manipulation bi-manual bimanual manipulation ✅
15 Fisheye-VLA: Decoupling Coverage and Acuity for Manipulation with a Single Fisheye Camera 提出Fisheye-VLA以解决单一鱼眼相机的操作视觉问题 manipulation VLA
16 Induced Riemannian Metrics for Motion Planning with Constraints 提出诱导黎曼度量以解决约束运动规划问题 manipulation bi-manual bimanual manipulation
17 RoboTwin-Phys: Do WAMs and VLAs Understand the Physical World? 提出RoboTwin-Phys以解决机器人操作中的物理条件多样性问题 manipulation physically plausible VLA
18 MAVP: Map-Aware Visuomotor Policies for Mobile Manipulation 提出MAVP框架以解决移动操控中的空间对齐问题 manipulation mobile manipulation ✅
19 Digital Twin-Driven VR Teleoperation with Multi-View Spatial Perception for Surgical Robots 提出数字双胞胎驱动的VR远程操作平台以解决手术机器人视角限制问题 teleoperation VR teleoperation
20 SparseNav: Instruction-conditioned Sparse Semantic Perception for Training-Free Vision-Language Navigation 提出SparseNav以解决视觉语言导航中的语义感知成本问题 quadruped Unitree open-vocabulary
21 DynaForge: Planning-Guided Residual Learning for Dynamic Manipulation Demonstration Generation 提出DynaForge以解决动态物体操控演示生成问题 manipulation sim-to-real reinforcement learning
22 φ-RIE: From Photorealistic Reconstruction to Interactive Environments 提出φ-RIE以解决3D场景交互性不足问题 manipulation 3D gaussian splatting 3DGS
23 A Deployable Four-Finger Payload for Teleoperated Free-Flying Manipulation with Astrobee 提出可部署的四指抓取装置以实现Astrobee的遥操作自由飞行 manipulation bi-manual teleoperation
24 IndustrialVLA-Bench: A Traceable Multi-Axis Evaluation of Open Robot Policy Models 提出IndustrialVLA-Bench以统一评估机器人政策模型的能力与鲁棒性 manipulation policy learning vision-language-action ✅
25 MotionForge: A Data Generation Pipeline and Large-Scale Benchmark for Long-Horizon Manipulation of Dynamic Objects with Domain Shifts 提出MotionForge以解决动态物体长时间操控中的评估与执行问题 manipulation embodied AI
26 PAKT: Physically-Aligned Kinesthetic Teaching for Reinforcement Learning 提出PAKT框架以解决工业操作中的强化学习指导问题 manipulation teleoperation reinforcement learning ✅
27 Wheel-loader V-Cycle Automation with Deep Koopman MPC 提出深度Koopman MPC以实现轮式装载机的V型循环自动化 MPC model predictive control
28 Hierarchical Floorplan-Guided Vision-Language Exploration for Embodied Question Answering 提出HFLEX-EQA以解决环境探索中的信息获取问题 quadruped semantic map open-vocabulary
29 Manipulation of Deformable Linear Objects Using Model Predictive Path Integral Control with Bidirectional Long Short-Term Memory Learning 提出基于双向LSTM的模型预测路径积分控制以解决可变形线性物体操控问题 manipulation model predictive control
30 Safety-Constrained Model Predictive Control for an Omnidirectional Walking Assistive Robot Using Control Barrier Function 提出基于控制障碍函数的模型预测控制以提升步态辅助机器人安全性 MPC model predictive control
31 PhyVisGen: Physically and Visually High-Fidelity Robotic Manipulation Data Generation 提出PhyVisGen以解决机器人操作数据生成的物理与视觉一致性问题 manipulation scene reconstruction
32 Relative Contact Velocity-Controlled Hand-Object Mechanism for Dexterous Tool Manipulation 提出统一手-物机制以实现灵巧工具操作 manipulation motion planning
33 What is the Better Curriculum: Controller-Shaped Grasping Behavior for Contact Force-Sensitive Manipulation 提出控制器形状抓取行为以解决接触力敏感操作问题 manipulation policy learning
34 From Instrument-Mounted Demonstrations to In-Vivo Execution: Learning Bimanual Laparoscopic Appendectomy Without Robot-Collected Demonstrations 提出一种新方法以实现无机器人收集演示的双手腹腔镜阑尾切除术 bi-manual diffusion policy
35 RoboMP-DINOv2: Prompts, Not Filters for Robust Robot Manipulation 提出RoboMP-DINOv2以解决机器人操作中的视觉泛化问题 manipulation diffusion policy ✅
36 A Reconfigurable Bidirectional Cable-Driven Hip Exoskeleton with Swappable Bench/Backpack Dual-configuration Actuation 提出可重构双向电缆驱动髋部外骨骼以解决便携性与功能性问题 motion tracking multimodal
37 Generalizing Manipulation Skills with a Local Coding Agent 提出本地编码代理以解决机器人任务泛化问题 manipulation
38 Manipulation with Stability Guarantees: Linear Deformable Objects with Non-negligible Physical Response Grasped at Multiple Location 提出闭环控制架构以解决可变形物体操控中的动态响应问题 manipulation
39 Predict Before You Step: Auditable Occupancy Forecasting for Dynamic Obstacle Avoidance under Sparse Guidance 提出LOOP以解决动态障碍物避让问题 legged robot locomotion Unitree
40 Learning Air-Ground Motion Control with Temporal Mode Switching and Cross-Terrain Tracking 提出基于学习的气地运动控制框架以解决多地形跟踪问题 locomotion dynamics randomization reinforcement learning
41 DreamStream: Towards Policy-Oriented Generative Simulation for End-to-End Driving 提出DreamStream以解决驾驶策略评估中的视觉差距问题 sim-to-real ✅
42 Beyond End-Task Success: How to Audit Visual Experience Retrieval in Robotics 提出审计方法以评估机器人视觉经验检索的有效性 manipulation
43 AgriGen: Large-Scale Scene Generation Framework for Photorealistic Agricultural Robotics Simulation 提出AgriGen框架以解决农业机器人模拟中的数据获取问题 domain randomization ✅

🔬 支柱三：空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗⭐
44 ArborSplat: Online Semantic Gaussian Splatting SLAM for Orchards 提出ArborSplat以解决果园机器人地图构建问题 monocular depth 3D gaussian splatting 3DGS
45 Dual Covariance Gaussian Splatting SLAM: Decoupling Rendering and Registration for Robust Real-Time Tracking 提出双协方差高斯点云SLAM以解决实时跟踪中的渲染与配准冲突问题 3D gaussian splatting 3DGS gaussian splatting
46 MOLA LiDAR-Inertial Odometry (MOLA-LIO) on the COMFORT Localization Benchmark 提出MOLA-LIO以提升LiDAR惯性测量精度 LIO
47 CDKF-Track: Cluster-aware Data-Driven Kalman Filtering for Cooperative 3D Multi-Object Tracking 提出CDKF-Track以解决3D多目标跟踪中的冗余测量问题 scene understanding

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
48 Beyond Reconstruction Error: Analytical and Data-Driven Action Tokenization for Autoregressive Vision-Language-Action Models 提出基于分析与数据驱动的动作标记化方法以优化自回归视觉-语言-动作模型 vision-language-action VLA
49 Destination Support Restoration for Finite-Set Multimodal Trajectory Prediction 提出目的支持恢复方法以解决多模态轨迹预测中的信息不足问题 multimodal
50 Zephyron: Integrated Design and Analytical Evaluation of a Solar-Assisted Mobile Manipulator for Multimodal Environmental Reconnaissance and Distributed Visual Inference 提出Zephyron以解决环境侦察中的能量与通信限制问题 multimodal
51 Deploying Foundation Models for Embodied Navigation 提出基于基础模型的导航方法以解决个性化与记忆管理问题 foundation model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
52 Skytopia: Monocular Drone Navigation with Action-Conditioned Latent World Models 提出Skytopia以解决单目无人机导航问题 world model world models 3D gaussian splatting
53 CableVLA: Simulation-Privileged Global-Local Representation Learning for Cable Routing 提出CableVLA以解决电缆布线中的全局与局部表示学习问题 representation learning vision-language-action multimodal
54 History-Conditioned Flow Matching for Probabilistic Dynamics of Tendon-Driven Continuum Robots 提出历史条件流匹配方法以解决腱驱动连续机器人动力学不确定性问题 flow matching

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
55 RoboFollow: Unveiling the Instruction Following Mirage in Embodied Agents 提出RoboFollow以揭示具身智能体的指令跟随能力不足问题 classifier-free guidance VLA instruction following ✅

⬅️ 返回 cs.RO 首页 · 🏠 返回主页