cs.RO（2026-09-18）

📊 共 57 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (43 🔗6) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱七：动作重定向 (Motion Retargeting) (3 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1)

🔬 支柱一：机器人控制 (Robot Control) (43 篇)

#题目一句话要点标签🔗⭐
1 FOCAL-VLA: Subtask-Guided Geometry Distillation and Implicit World Modeling for Vision-Language-Action Models 提出FOCAL-VLA以解决视觉-语言-动作模型的几何理解不足问题 manipulation world model world models ✅
2 A Sim-to-Real Integration Pipeline for Training and Deployment of Chunk-Based VLA Manipulation Policies 提出一种Sim-to-Real集成管道以解决VLA操作策略训练问题 manipulation sim-to-real sim2real
3 Skel-WAM: A Hand-Skeleton-Conditioned World Action Model for Human-to-Robot Manipulation Transfer 提出Skel-WAM以解决人机操作转移中的动作差异问题 manipulation bi-manual world action model
4 SABER: Learning Attention-based Semantic Affordance for Legged Locomotion 提出SABER以解决腿部机器人在复杂地形中的接触安全问题 legged robot legged locomotion locomotion
5 Learning Distance-Conditioned Object Transport for Humanoid Loco-Manipulation from a Single Motion Clip 提出距离条件物体运输学习以解决人形机器人运动再现问题 humanoid motion tracking manipulation
6 SafeStage: Evaluating Safety Before, During, and After Vision-Language-Conditioned Robot Manipulation 提出SafeStage以评估视觉语言条件下机器人操作的安全性 manipulation world model world models
7 SynthDemo-RL: Breaking the Zero-Reward Barrier in VLA Adaptation with LLM-Guided Synthetic Demonstrations 提出SynthDemo-RL以解决VLA适应中的零奖励问题 manipulation teleoperation reinforcement learning
8 FootQuery: Future-Touchdown-Guided Retrieval from Depth History for Perceptive Humanoid Locomotion 提出FootQuery以解决复杂地形下人形机器人行走问题 humanoid humanoid locomotion locomotion
9 PSR: Predictive Sensorimotor Representation Learning for Contact-Rich Manipulation 提出预测传感运动表征学习以解决接触丰富操控问题 manipulation representation learning vision-language-action
10 SkelWAM: A Skeleton-Guided World-Action Model for Zero-Shot Cross-Embodiment Manipulation 提出SkelWAM以解决跨体态操控的经验重用问题 manipulation cross-embodiment embodiment transfer
11 CommitFlow: Semantic Commitment Verification and Local Correction for Long-Horizon Robot Manipulation VLA Execution 提出CommitFlow以解决长时间机器人操作中的语义承诺验证问题 manipulation vision-language-action VLA
12 Potential-Field Action Representation for Reinforcement Learning in Contact-Rich Manipulation 提出PA-RL框架以解决接触丰富操作中的动作表示问题 manipulation reinforcement learning motion generation
13 FAN: Foresight Action Normalization for Continual Adaptation of Vision-Language-Action Models 提出FAN以解决视觉-语言-动作模型的持续适应问题 manipulation bi-manual bimanual manipulation
14 Duty Factor Predicts Robust Constrained Quadrupedal Locomotion Across Gait Types 提出以占空比为基础的四足机器人稳健运动控制方法 quadruped locomotion model predictive control
15 Contact-Rich Motion Planning via GPU-Parallel Mode Evaluation 提出CoMET以解决接触丰富运动规划的计算挑战 locomotion manipulation bi-manual
16 Outcome-Conditioned End-Effector Geometry Across Vision-Language-Action Policies 研究跨策略末端执行器几何以提升视觉语言动作策略的成功率 manipulation vision-language-action VLA
17 DPed-VLN: A Benchmark for Socially Compliant Vision-and-Language Navigation in Dynamic Pedestrian Environments 提出DPed-VLN以解决动态行人环境中的视觉语言导航问题 humanoid reinforcement learning imitation learning
18 Robotic Multiphase Interaction: Manipulating Coupled Liquid and Solid Dynamics with a World Model 提出机器人多相交互以解决液体与固体动态操控问题 manipulation diffusion policy flow matching
19 When to Waddle: A Comparative Study of Bipedal Torso-Stabilization on Low-Friction Surfaces 提出基于企鹅摇摆的双足稳定策略以应对低摩擦表面挑战 bipedal biped locomotion
20 AirSplan: Risk-Aware Motion Planning for Quadrotors in Cluttered 3D Gaussian Splats 提出AirSplan以解决复杂环境中四旋翼的安全运动规划问题 motion planning 3D gaussian splatting gaussian splatting
21 Safe Real-Time Policy Steering via Noise-Space Trajectory Optimization for One-Step Generative Policies 提出INSPO以解决生成策略实时控制中的约束适应问题 trajectory optimization vision-language-action multimodal
22 Sandwich-Residuals: Parameter-Efficient Test-time Adaptation of World Models 提出Sandwich-Residuals以解决世界模型测试时适应性问题 manipulation world model world models
23 Adaptive Rollout Truncation Based on Epistemic Uncertainty for Efficient Offline World Model Training 提出基于认知不确定性的自适应截断策略以提高离线世界模型训练效率 ANYmal world model world models
24 ZeroTouch: Tactile-Supervised Visual Contact Estimation for Contact-Rich Manipulation 提出ZeroTouch框架以解决机器人抓取中的接触估计问题 manipulation MAE OpenVLA
25 KnowDemo: Knowledge-Guided Robot Demonstration Generation from Human Videos 提出KnowDemo以解决机器人演示生成的效率与多样性问题 manipulation sim-to-real motion planning ✅
26 LIMBO: Learning and Internalizing Model-Free Barrier Objectives for Agile and Safe Whole-Body Control 提出LIMBO框架以解决安全全身控制问题 humanoid whole-body control locomotion
27 Stochastic Neural Signed Swept Volume for Real-time Chance-Constrained Trajectory Optimization 提出随机神经签名扫掠体积以解决实时机会约束轨迹优化问题 manipulation trajectory optimization motion planning
28 MAAP: Multi-Agent Active Perception for Collaborative Manipulation 提出MAAP以解决多智能体协作操作中的感知不足问题 manipulation dual-arm imitation learning
29 From Pretraining to Proficiency: Real-World Subtask RL for Long-Horizon Manipulation with Minimal Human Intervention 提出PARTS框架以解决长时间操作中的子任务强化学习问题 manipulation bi-manual reinforcement learning
30 ProTracer: Proprioception-Guided Failure Diagnosis in Robot Manipulation 提出ProTracer以解决机器人操作中的故障诊断问题 manipulation multimodal
31 SeeQ: Training Generalist Value Functions for Long-Horizon Robotic Manipulation 提出SeeQ以解决长时间机器人操作中的价值函数学习问题 manipulation bi-manual
32 Towards Fine-Grained Object Manipulation: SAM3-Guided Visuomotor Policy with Persistent Memory Learning and Focused Visual Conditioning 提出SAM3引导的视觉运动策略以解决细粒度物体操控问题 manipulation diffusion policy
33 AcousticDiffusion: Semantically Conditioned Audio-Guided Diffusion Policy for Search-and-Rescue Assistance 提出AcousticDiffusion以解决救援机器人导航问题 quadruped diffusion policy
34 LEMCA: LLM-Guided Synthesis of Efficient Mode-Switching Control Architectures 提出LEMCA以解决机器人控制模式切换效率问题 manipulation reinforcement learning large language model
35 Stability-aware Residual Reinforcement Learning Framework for Robotic Manipulator Disturbance Compensation 提出残差强化学习框架以解决机器人操控中的干扰补偿问题 sim-to-real reinforcement learning
36 Scaling Vision-Language Reward Learning for Robot Manipulation in Parallel Simulation 提出RAPID以加速机器人操作中的视觉-语言奖励学习 manipulation ✅
37 ForceTwin: Physics-informed Digital Twins for Robotic Manipulation from Instrumented Human Interaction 提出ForceTwin以解决机器人操作中的物理特性识别问题 manipulation ✅
38 NeuRIO: A Streaming Neural Estimator for Zero-Shot Sim-to-Real Multi-Robot Relative Inertial Odometry 提出NeuRIO以解决多机器人相对惯性测距问题 sim-to-real
39 Gripper-Aware Automatic Dense Packing of Irregular Objects 提出一种抓手感知的自动密集包装方法以解决不规则物体包装问题 manipulation height map
40 FORTE: Task-Adaptive Force Capability Optimization for Mobile Manipulators 提出任务自适应力能力优化框架以提升移动操控器性能 manipulation trajectory optimization ✅
41 CRISP: Contact-Rich Robotic Simulation Platform with Extensive Geometries and Contact Solvers 提出CRISP以解决复杂多接触机器人仿真问题 manipulation ✅
42 A High-Payload Wall-Climbing Robot Using Passive Bistable Suction Cups 提出一种高载重的被动双稳态吸附机器人以解决墙面攀爬问题 locomotion
43 LOInK: Learned Optimal Inverse Kinematics via Structured Neural Surrogate Models 提出LOInK以高效解决逆向运动学问题 quadruped

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
44 2nd Place Solution to the HANDS 2026 Workshop Challenge-Dexterous Grasp Motion Track: Single-Shot Trajectory Warping for Grasp Motion Generation 提出单次轨迹变形方法以解决灵巧抓取运动生成问题 PPO motion generation physically plausible
45 Fewer Steps, Better Actions: Rethinking Flow-Matching Inference for VLA Policies 提出Coda以优化流匹配推理，提升VLA策略性能 flow matching vision-language-action VLA
46 Compact but Moving: Intervention-Relevant Geometry in Recurrent World Models 提出高维动态中的紧凑干预几何结构以解决模型响应问题 world model world models
47 NaViRrator: Robot Navigation from Human-Readable Maps through a Learned Visual Route 提出NaViRrator以解决机器人导航与人类可读地图的连接问题 flow matching egocentric VLN
48 CARF: Contrastive Attraction-Repulsion of Failure-Guided Flow Matching 提出CARF框架以解决机器人演示数据中的失败轨迹问题 flow matching ✅
49 MicroHookACT: Monocular Microscopic Vision Guided Visuomotor Policy for Flexible Microelectrode Hooking 提出MicroHookACT以解决灵活微电极植入中的自动化钩针问题 policy learning imitation learning

🔬 支柱七：动作重定向 (Motion Retargeting) (3 篇)

#题目一句话要点标签🔗⭐
50 GALA: Geometry-Aware Latent Action Modeling for Vision-Language-Action Model Pretraining across Embodiments 提出GALA框架以解决多体现数据中的动作表示问题 cross-embodiment motion representation vision-language-action ✅
51 The EventCV Library for Event-Based Robotic Vision 提出EventCV库以简化事件相机在机器人视觉中的应用 motion estimation
52 Visual Proactivity: Enhancing Human-Robot Collaboration Through Intent Communication 提出视觉主动性以增强人机协作中的意图沟通 human-to-robot

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
53 VLA-Scope: Shift-Aware Failure Prediction for Vision-Language-Action Models 提出VLA-Scope以解决视觉-语言-动作模型的失败预测问题 vision-language-action VLA OpenVLA
54 AVT-Fabric: Active Visuo-Tactile Perception via Adaptive Evidence Selection for Efficient Robotic Fabric Comparison 提出AVT-Fabric以解决机器人面料比较中的视觉与触觉融合问题 large language model multimodal
55 When Should Robots Intervene? Balancing Engagement and Intrusiveness in Human-Robot Interaction 提出基于上下文感知的干预策略以优化人机交互体验 multimodal

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
56 AtomEgo: Exploring Ego-Robot Integration for Embodied Foundation Model Pretraining 提出AtomEgo以解决机器人与人类交互数据整合问题 egocentric cross-embodiment foundation model

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

#题目一句话要点标签🔗⭐
57 Robust Structureless Monocular Visual Inertial Initialization Exploiting Line Features and Vanishing Points 提出SLIM-init以解决视觉惯性里程计初始化不稳定问题 VIO ✅

⬅️ 返回 cs.RO 首页 · 🏠 返回主页