cs.RO（2026-09-29）

📊 共 85 篇论文 | 🔗 15 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (62 🔗11) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (8 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (4 🔗1) 支柱四：生成式动作 (Generative Motion) (1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱一：机器人控制 (Robot Control) (62 篇)

#题目一句话要点标签🔗⭐
1 EgoHumanoid-V2: Human-to-Humanoid Transfer of Coordinated Whole-Body Skills for Loco-Manipulation 提出EgoHumanoid-V2以解决人类技能向类人机器人转移问题 humanoid manipulation loco-manipulation
2 FineART: Fine-grained Annotated Robotic Trajectory Dataset and Vision-Language-Action Model for Bimanual Manipulation 提出FineART数据集与模型以解决双手操作任务的复杂性问题 manipulation bi-manual bimanual manipulation
3 Geometry-Preserving Human-to-Robot Upper-Body Motion Retargeting from Monocular Video 提出几何保持的人机上半身动作重定向框架以解决视频驱动的挑战 dual-arm sam 3D SAM 3D
4 AeroManip-VLA: Scalable Vision-Language-Action Learning for Aerial Manipulation with RL-Generated Demonstrations 提出AeroManip-VLA以解决无人机操作中的数据生成与评估问题 manipulation teleoperation reinforcement learning
5 Learning Expressive and Compositional Motion Representation via Spectral Skills 提出光谱技能以解决人形机器人控制中的行为组合问题 humanoid humanoid robot Unitree
6 EquivDP3: A SIM(3)-Invariant Point-Cloud Encoder for Data-Efficient Humanoid Loco-Manipulation 提出EquivDP3以解决人形机器人低数据效率问题 humanoid locomotion locomotion policy
7 EgoAlign: Bridging the Human-Humanoid Gap for Long-Range Loco-Manipulation 提出EgoAlign以解决人类与类人机器人间的长距离运动操控问题 humanoid whole-body control locomotion ✅
8 OTRetarget: Joint Robot and Object Motion Retargeting via Optimal Transport 提出OTRetarget以解决机器人与物体运动重定向问题 humanoid humanoid robot manipulation
9 Counterfactual Video Generation Enables Scalable Humanoid Loco-Manipulation 提出PRISM框架以解决人形机器人运动操控技能训练问题 humanoid manipulation loco-manipulation
10 CrossBFM: Distilling a Shared Latent Behavior Space Across Humanoid Embodiments 提出CrossBFM以解决人形机器人行为空间共享问题 humanoid whole-body control motion tracking ✅
11 Rho: A Foundation for Efficiently Adaptable VLA Models 提出Rho模型以实现高效适应的双臂机器人操作 manipulation bi-manual dual-arm
12 Video2STL: Grounding VLM-Generated Temporal Specifications for Robot Learning 提出Video2STL框架以解决机器人学习中的视频信息转移问题 quadruped locomotion manipulation
13 Predictive Safety Curricula for Robust Legged Locomotion 提出预测安全课程以提升腿部机器人运动的可靠性 legged robot legged locomotion locomotion
14 Disentangling Spurious Correlations in Vision-Language-Action Models via Predicting Domain-Invariant Latent Lookahead 提出DILL框架以解决视觉-语言-动作模型中的虚假关联问题 manipulation policy learning representation learning ✅
15 Foundation-Model-Guided Topology-Aware Semantic Risk Fields for Manipulation 提出拓扑感知语义风险场以提升机器人操作安全性 manipulation motion planning foundation model
16 Urgent Actions Go First: Urgency-Aware Denoising for Real-Time VLA Control 提出紧急性意识去噪方法以解决实时VLA控制延迟问题 manipulation flow matching vision-language-action
17 DROM: A Language-Guided Diffusion Framework for Multi-Skill Robotic Manipulation 提出DROM框架以解决多技能机器人操作中的数据不足问题 manipulation motion planning behavior cloning ✅
18 Remember What You Did: Action-History Memory with Dual-Expert Denoising for Long-Horizon Vision-Language-Action Policies 提出ActMem-VLA以解决长时间任务中的动作歧义问题 manipulation Mamba vision-language-action
19 Cooperative Multi-Agent Vision-Language-Action Models via Reinforced Fine Tuning 提出三阶段强化微调方法以解决多智能体视觉-语言-动作模型协作问题 manipulation reinforcement learning vision-language-action
20 MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation 提出MotorMind以解决零-shot机器人操作的局限性 manipulation vision-language-action VLA
21 Faster and Better? Benchmark Bugs and Design Limitations Distort the Evaluation of Vision-Language-Action Acceleration 揭示基准缺陷与设计局限影响视觉-语言-动作加速评估 manipulation vision-language-action VLA
22 Beyond Token Importance: Preserving Spatial Scaffolds for Efficient Vision-Language-Action Inference 提出GeoScaffold以解决视觉语言行动推理中的空间信息丢失问题 manipulation vision-language-action VLA
23 Kinematic Nonlinear Spatio-Temporal Trajectory Warping for Contact-Rich Dexterous Manipulation Demonstrations 提出非线性时空轨迹变换方法以改善接触丰富的灵巧操作演示 manipulation dexterous hand dexterous manipulation
24 HACo: Learning Haptic Active Compliance for Force-Aware Dexterous Manipulation 提出HACo以解决力感知灵巧操作中的合规性问题 manipulation dexterous manipulation teleoperation
25 CogWAM: Aligning Semantic Cognition with World Action Modeling via Event-Driven Interfaces 提出CogWAM以解决机器人任务进展与行动预测不一致问题 manipulation dual-arm world action model
26 WorldLine: Action-Driven Visual Simulation for Robotic Manipulation 提出WorldLine以解决机器人操控中的视觉模拟问题 manipulation shared control distillation ✅
27 BlenDAgger: Blended Shared Control for Interactive Imitation Learning 提出BlenDAgger以解决机器人模仿学习中的人类干预效率问题 manipulation shared control imitation learning
28 VidAct: Learning Manipulation from In-the-Wild Videos with Object-Centric 3D Awareness 提出VidAct以解决视频学习操控中的3D几何感知问题 manipulation sim-to-real policy learning
29 World4Scorer: Outcome-Grounded World Modeling for Autonomous Driving 提出World4Scorer以解决自主驾驶中的轨迹评分问题 manipulation world model world models
30 RawVLA: Embodied Neural Image Signal Processor For Robotic Manipulation 提出RawVLA以解决视觉-语言-动作模型中的图像处理问题 manipulation vision-language-action VLA
31 FP2: Equipping Robotic Foundation Models with Force Control 提出FP2以解决机器人基础模型在力控制中的挑战 manipulation foundation model ✅
32 Multifunctional Locomotion Control of Multi-Jointed BURs with Swimming and gait Capabilities 提出多关节水下机器人多功能运动控制机制以解决行为选择限制问题 legged locomotion locomotion multimodal ✅
33 TaRL: Learning General and Physical Rewards from Tactile Demonstrations 提出TaRL框架以从触觉示范中学习奖励 manipulation reinforcement learning policy learning ✅
34 T$^2$Mem: Learning Test-Time Memory for Robotics 提出T$^2$Mem以解决机器人记忆依赖问题 manipulation policy learning vision-language-action ✅
35 ComManip: Overfitting Manipulation Policies to Comfortable Regions 提出ComManip以解决机器人操作策略泛化不足问题 manipulation OpenVLA
36 Trajectory-Level Mode Guidance for Controllable Diffusion-Based Multi-Robot Motion Planning 提出基于轨迹级模式引导的可控扩散多机器人运动规划方法 motion planning multimodal
37 In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks 提出简单化的机器人上下文学习框架以解决任务执行不明确问题 manipulation affordance ✅
38 Encore: Few-Shot Agentic Discovery of Manipulation Strategies 提出ENCORE以解决机器人任务中的策略发现问题 manipulation bi-manual
39 Recovering the View: Benchmarking Physical Active Vision for Occlusion Recovery in Robotic Manipulation 提出BAVO-Bench与A-FAR以解决机器人操作中的遮挡恢复问题 manipulation bi-manual
40 PreferenceFlow: Test-Time Guidance of Flow-Matching Robot Policies from Human Interventions 提出PreferenceFlow以解决机器人策略在测试时的引导问题 manipulation reinforcement learning flow matching
41 Staircase Policy: Streaming Inference for World-Action Models with Large Action Chunks 提出阶梯策略以解决世界-动作模型的推理效率问题 manipulation flow matching JEPA
42 Learning to Explore Hidden Kinematics for Articulated Object Manipulation 提出一种新方法以解决关节物体操作中的运动学模糊问题 manipulation reinforcement learning
43 A robust single-sensing-element tactile sensor for concurrent pressure and tackiness detection with real-time signal decoupling capability 提出一种新型触觉传感器以解决多模态信号解耦问题 humanoid humanoid robot multimodal
44 ReF-HIL: Shaping the Critic around Human Action Neighborhoods for Efficient Human-in-the-Loop Reinforcement Learning 提出ReF-HIL框架以解决人机协作强化学习效率问题 manipulation reinforcement learning
45 FORM: Robot Manipulation through Direct Material Law Identification 提出FORM以解决机器人对未知可变材料的操控问题 manipulation
46 MVG-WAM: Multiple View Geometry-Aware World-Action Modeling for Robotic Manipulation 提出MVG-WAM以解决多视角几何关系不足的问题 manipulation
47 RoboHarn-Evo: Evolving Hierarchical Physical Knowledge for Self-Improving Robotic Manipulation 提出RoboHarn-Evo以解决机器人操控中的物理知识不足问题 manipulation
48 RoboFin3D: A Sim-to-Real Platform for Robotic Surface Finishing 提出RoboFin3D以解决机器人表面抛光中的高成本实验问题 sim-to-real
49 All You Need Is Low Fidelity: Zero-Shot Sim-to-Real of Learned Robotic Fish Control 提出低保真度模拟器以实现机器人鱼的零-shot控制 sim-to-real
50 Track-and-Complete: Learning Humanoid Skills from a Single Failed Human Video 提出TRACC以从失败人类视频中学习类人技能 humanoid
51 Wrench-ACT: Enhancing Robot Policies for Contact Rich Behavior Using Direct Wrench Control 提出Wrench-ACT以解决机器人接触丰富行为的控制问题 manipulation teleoperation imitation learning
52 Credit-Guided Policy Improvement for Test-time Adaptive Vision-Language Navigation 提出信用引导的策略改进以解决测试时自适应视觉语言导航问题 sim-to-real VLN
53 DRHeC: Differentiable Rendering for Hand-Eye Calibration with RGB-Based Gradients 提出RGB基础的可微渲染框架以解决手眼标定精度问题 manipulation geometric consistency
54 Inferring Soil Friction Angle from Robot Foot-Ground Force Histories: A Bayesian Inverse Approach to Proprioceptive Soil Sensing 提出基于贝叶斯逆推的土壤摩擦角推断方法 quadruped world model world models
55 Recompositional Robotics: Cross-Domain, Open-set, and Lifelong Modularity Beyond Morphology 提出跨域重组机器人以解决能力适应性问题 motion planning
56 ProAct-VLM: Pre-Failure Vision-Language Task Replanning with Continuous Perception Feedback 提出ProAct-VLM以解决动态环境下机器人任务重规划问题 manipulation ✅
57 From Sky to Soil: A Morphing Aerial-Ground Robot for Seed Deployment 提出一种变形空地机器人以解决种子播撒精度不足问题 locomotion
58 RoboChrono: A Real Robot Benchmark for Streaming Task Understanding 提出RoboChrono基准以解决机器人任务理解问题 manipulation
59 Executor-aware Candidate Selection via a Feasibility Certificate 提出基于证书的候选选择框架以解决执行器兼容性问题 motion planning
60 Simple Agentic Memory for Generalist Robot Policies 提出简单代理记忆以解决通用机器人策略的记忆依赖问题 manipulation
61 Riemannian Splat Regression Models for Learning Time Fields on Arbitrary Riemannian Manifolds 提出无坐标投影的Riemannian Splat回归模型以学习时间场 motion planning
62 DQ-MPCC: Dual-Quaternion MPCC for Quadrotor Racing 提出双四元数MPCC以解决四旋翼竞速控制问题 sim-to-real

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
63 V2X-WAM: A Cooperative World Action Model for End-to-End Autonomous Driving 提出V2X-WAM以解决未来交通环境预测不足的问题 world action model world action models scene understanding
64 Anisotropic Representations Improve Planning in JEPA World Models 提出AnisoWM以解决任务对齐的潜在规划成本问题 world model worldmodel world models ✅
65 Where Predictive Supervision Goes Shapes What VLA Policies Learn 提出预测监督机制以提升视觉语言行动策略的表现 representation learning vision-language-action VLA
66 One from Infinity: Actualizing Futures from Pretrained World Models into Robot Actions 提出RoboActualizer以解决机器人动作执行中的任务条件选择问题 flow matching world model world models
67 PhysWAM: Physically Consistent World Action Model for Autonomous Driving 提出PhysWAM以解决自主驾驶中的世界与行动建模问题 flow matching world action model world action models
68 Distilling Privileged Control Barrier Functions into RGB-Only Safety Filters for Dynamic Visual Navigation 提出教师-学生蒸馏框架以解决动态环境中的RGB导航安全问题 teacher-student distillation 3D reconstruction ✅
69 Reactive Real-Time Flow Policies via Asynchronous Distribution Alignment 提出异步分布对齐方法以提升机器人实时反应能力 distillation vision-language-action VLA
70 All Roads Lead to Rome: Flow-driven Multi-Anchor Exploration for Open-Environment Active 3D Mapping 提出条件流匹配以解决开放环境主动3D映射中的探索问题 flow matching multimodal
71 Planning Oriented 3D Scene Completion via Coupled TUDF Occupancy Representation Learning from Partial Observations 提出面向路径规划的3D场景补全框架以解决部分可观测性问题 representation learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗⭐
72 Taming VLAs under Robot Execution Errors: Self-Compensation and Stress Testing 提出自补偿VLA以解决机器人执行错误问题 vision-language-action VLA
73 doPlan: A Variable-Horizon Dataset for Multi-Stage Language-Conditioned Planning in Autonomous Driving 提出doPlan数据集以解决多阶段语言条件规划问题 language conditioned ✅
74 WayFinder: Hierarchical Visual-Language-Action for Zero-Shot Waypoint Generation and Low-Level Kinematic Control 提出WayFinder以解决自主机器人导航中的执行不可靠性问题 VLA large language model multimodal
75 CoRe-VLA: Preserving Cross-View Coordination in VLAs under Camera Shifts 提出CoRe-VLA以解决相机位移下的跨视角协调问题 VLA
76 Explore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents 提出RoboSkill框架以解决机器人任务泛化问题 vision-language-action multimodal
77 LexiconVLA: Learning Reusable Atomic Action Codebooks for Unseen Tasks 提出LexiconVLA以解决未见任务中的动作重用问题 vision-language-action VLA
78 Skill-Space Shooting for Autonomous Robot Policy Improvement 提出技能空间射击以提升自主机器人策略 foundation model
79 Risk-Aware Semantic Grounding for Trustworthy LLM-Based Robot Planning 提出风险感知语义基础框架以解决LLM机器人规划的信任问题 large language model

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

#题目一句话要点标签🔗⭐
80 When to Adapt: Multi-Signal Domain Shift Detection for Efficient Training-Free Adaptation in Open-Vocabulary Segmentation 提出多信号域偏移检测方法以实现高效的无训练适应 open-vocabulary open vocabulary foundation model
81 Semantic Map Sharing and Capability-Aware Coverage Planning for AI-Native 6G Robotic Coordination 提出语义地图共享与能力感知覆盖规划以解决SAR机器人协调问题 semantic map traversability
82 Pow3R-SLAM: Real-Time RGB-D SLAM with 3D Reconstruction Priors 提出Pow3R-SLAM以解决RGB-D SLAM中的深度稀疏问题 3D reconstruction ✅
83 Asymmetric Scout-Worker Reconnaissance for Route Validation in Unknown Environments 提出不对称侦察-工作者框架以解决未知环境中的路径验证问题 traversability

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
84 Comparing Utility of Inertial, Occupancy, Semantic, and Intent Information in Human Motion Prediction During Daily Tasks 提出人类运动预测模型以提升室内导航精度 motion diffusion model motion diffusion human motion

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

#题目一句话要点标签🔗⭐
85 Scale-Invariant Manipulability Shape Tracking Across Heterogeneous Manipulators 提出一种尺度不变的可操控性形状跟踪方法以解决异构机械臂间的转移问题 human-to-robot

⬅️ 返回 cs.RO 首页 · 🏠 返回主页