cs.RO（2026-06-25）

📊 共 36 篇论文 | 🔗 5 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) (4) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱四：生成式动作 (Generative Motion) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱一：机器人控制 (Robot Control) (27 篇)

#题目一句话要点标签🔗⭐
1 In-Context Model Predictive Generation: Open-Vocabulary Motion Synthesis from Language Models to Physics 提出ICMPG框架以解决人类动作合成中的语义与物理现实性问题 MPC model predictive control open-vocabulary
2 HumanoidUMI: Bridging Robot-Free Demonstrations and Humanoid Whole-Body Manipulation 提出HumanoidUMI框架以解决机器人演示数据收集效率低下问题 humanoid humanoid robot whole-body control
3 PressMimic: Pressure-Guided Motion Capture and Control for Humanoid Robot Imitation 提出PressMimic以解决类人机器人模仿中的接触动态问题 humanoid humanoid robot humanoid control
4 PhysReflect-VLA: Physical Feasibility and Self-Reflective Regulation for Reliable Vision-Language-Action Policies 提出PhysReflect-VLA以解决长时间机器人操作中的物理可行性问题 manipulation vision-language-action VLA
5 PAMAE: Phase-Aware-MoE Action Experts Towards Reliable Flow-Matching Vision-Language-Action Policies 提出PAMAE以解决多阶段机器人操作中的可靠动作生成问题 manipulation flow matching vision-language-action
6 Humanoid-DART: Humanoid Loco-Manipulation using Diffusion-guided Augmentation through Relabeling and Tracking 提出自监督框架以解决人形机器人运动操控中的示范收集挑战 humanoid manipulation loco-manipulation
7 Improving Vision-Language-Action Model Fine-Tuning with Structured Stage and Keyframe Supervision 提出StaKe框架以解决VLA模型微调中的结构化监督问题 manipulation bi-manual vision-language-action ✅
8 World Action Models Enable Continual Imitation Learning with Recurrent Generative Replays 提出递归生成重放框架以解决机器人持续模仿学习问题 manipulation imitation learning world action model
9 RelAfford6D: Relational 6D Affordance Graphs for Constraint-Driven Robotic Manipulation 提出RelAfford6D以解决复杂机器人操控中的约束问题 manipulation affordance foundation model
10 Scalable Behavior Cloning with Open Data, Training, and Evaluation 提出ABC以解决行为克隆中的数据获取与训练问题 manipulation teleoperation behavior cloning
11 SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation 提出SSI-Policy以解决低数据环境下的机器人视觉语言操控问题 manipulation monocular depth cross-embodiment
12 VibeAct: Vibration to Actions for Contact-Rich Reactive Robot Dexterity 提出VibeAct以解决接触丰富环境下机器人灵巧操作问题 manipulation dexterous hand dexterous manipulation ✅
13 Learning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline) 提出基于强化学习的双手衣物折叠解决方案 bi-manual sim-to-real reinforcement learning
14 LA4VLA: Learning to Act without Seeing via Language-Action Pretraining 提出LA4VLA框架以解决视觉依赖问题 manipulation vision-language-action VLA
15 Tactile-WAM: Touch-Aware World Action Model with Tactile Asymmetric Attention 提出Tactile-WAM以解决接触丰富操作中的视觉预测不足问题 manipulation world action model world action models
16 Bridging Handheld and Teleoperated Supervision for Contact-Rich Manipulation via State-Gated Experts 提出BRIDGE方法以解决手持与遥控监督在接触丰富操作中的融合问题 manipulation teleoperation diffusion policy
17 E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulation 提出E-TTS框架以解决机器人操作中的测试时间缩放问题 manipulation vision-language-action
18 Advancing Omnimodal Embodied Agents from Isolated Skills to Everyday Physical Autonomy 提出OmniAct框架以解决机器人自主性与工具整合问题 manipulation VLA multimodal
19 IDEA: Insensitive to Dynamics Mismatch via Effect Alignment for Sim-to-Real Transfer in Multi-Agent Control 提出一种基于效应对齐的多智能体控制方法以解决动态不匹配问题 sim-to-real policy learning
20 Learning Motion Feasibility from Point Clouds in Cluttered Environments 提出从点云学习运动可行性以解决复杂环境中的规划问题 manipulation motion planning task and motion planning
21 BOWConnect: Parallel Bayesian Optimization over Windows with Learned Local Cost Maps for Sample-Efficient Kinodynamic Motion Planning 提出BOWConnect以解决高维运动规划中的样本效率问题 motion planning ✅
22 RobOralScan: Learning Active Intraoral Scanning for Robotic Dental Reconstruction 提出RobOralScan以解决口腔内扫描自动化不足问题 sim-to-real reinforcement learning
23 SceneBot: Contact-Prompted General Humanoid Whole Body Tracking with Scene-Interaction 提出SceneBot以解决人形机器人在接触丰富环境中的全身运动跟踪问题 humanoid humanoid control motion tracking ✅
24 LA4VLA: Learning to Act without Seeing via Language-Action Pretraining 提出LA4VLA框架以解决视觉-语言-动作模型的依赖问题 manipulation vision-language-action VLA
25 Learning to Throw: Agile and Accurate Cable-Suspended Payload Delivery with a Quadrotor 提出混合仿真框架以解决悬挂载荷投放精度不足问题 manipulation trajectory optimization reinforcement learning
26 Support-Constrained RL Enables Real-World Policy Improvement without Real-World Experience 提出支持约束强化学习以解决现实世界策略改进问题 manipulation sim-to-real reinforcement learning ✅
27 AO-ARC: Almost-Surely Asymptotically Optimal Multi-Robot Motion Planning with ARC 提出AO-ARC以解决多机器人运动规划问题 motion planning

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

#题目一句话要点标签🔗⭐
28 ForesightSafety-VLA: A Unified Diagnostic Safety Benchmark for Vision-Language-Action Models 提出ForesightSafety-VLA以解决视觉-语言-动作模型的安全性问题 vision-language-action VLA
29 RouterVLA: Turning Smoke Tests into Supervision for Heterogeneous VLA Selection 提出RouterVLA以优化异构VLA策略选择 vision-language-action VLA
30 Proposal-Conditioned Latent Diffusion for Closed-Loop Traffic Scenario Generation 提出基于提案条件的潜在扩散以解决闭环交通场景生成问题 multimodal
31 Inference-Time Robot Behavior Steering through Physically-Aware Reconfiguration of Task-Structure 提出ReStruct以解决机器人政策推理时行为引导问题 VLA

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

#题目一句话要点标签🔗⭐
32 Risk-Aware Selective Multimodal Driver Monitoring with Driver-State World Modeling 提出成本感知选择性推理框架以解决驾驶员监控中的安全性问题 world model world models multimodal
33 Continual Robot Policy Learning via Variational Neural Dynamics 提出持续机器人策略学习框架以应对动态变化问题 policy learning differentiable simulation
34 PlanRL: A Trajectory Planning Architecture for Reinforcement Learning-based Driving Experts 提出PlanRL以解决RL驾驶专家的可解释性与规划复杂性问题 reinforcement learning policy learning

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
35 LAMP: Lane-Aligned Motion Primitives for Feasible Trajectory Prediction 提出LAMP以解决多模态轨迹预测中的拓扑一致性问题 VQ-VAE spatiotemporal multimodal

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

#题目一句话要点标签🔗⭐
36 OSC2Runner: OpenSCENARIO 2.x Compliant High-Fidelity AV Simulation in CARLA 提出OSC2Runner以解决OpenSCENARIO 2.x执行支持问题 spatiotemporal

⬅️ 返回 cs.RO 首页 · 🏠 返回主页