| 1 |
Learning Sim-to-Real Humanoid Locomotion in 15 Minutes |
提出FastSAC和FastTD3，15分钟内完成人形机器人Sim-to-Real强化学习 |
humanoid humanoid control humanoid locomotion |
|
|
| 2 |
ManualVLA: A Unified VLA Model for Chain-of-Thought Manual Generation and Robotic Manipulation |
提出ManualVLA，用于生成式操作手册和机器人操作，提升长时程任务性能。 |
manipulation 3D gaussian splatting gaussian splatting |
|
|
| 3 |
Modality-Augmented Fine-Tuning of Foundation Robot Policies for Cross-Embodiment Manipulation on GR1 and G1 |
提出模态增强微调框架，提升基座机器人策略在不同人形机器人上的操作能力。 |
humanoid manipulation motion planning |
|
|
| 4 |
Discovering Self-Protective Falling Policy for Humanoid Robot via Deep Reinforcement Learning |
提出基于深度强化学习的人形机器人自保护倒地策略 |
quadruped humanoid humanoid robot |
|
|
| 5 |
GR-RL: Going Dexterous and Precise for Long-Horizon Robotic Manipulation |
GR-RL：通过强化学习提升通用视觉-语言-动作策略在灵巧操作中的性能 |
manipulation dexterous manipulation reinforcement learning |
|
|
| 6 |
Learning Dexterous Manipulation Skills from Imperfect Simulations |
提出DexScreW框架，解决灵巧操作中模拟到真实的迁移难题，实现螺母螺栓紧固和螺丝刀拧紧。 |
manipulation dexterous manipulation sim-to-real |
|
|
| 7 |
DiG-Flow: Discrepancy-Guided Flow Matching for Robust VLA Models |
DiG-Flow：通过差异引导的Flow Matching提升VLA模型在机器人操作任务中的鲁棒性 |
manipulation flow matching vision-language-action |
|
|
| 8 |
A Cross-Embodiment Gripper Benchmark for Rigid-Object Manipulation in Aerial and Industrial Robotics |
提出跨平台机械爪基准测试CEGB，用于评估空中和工业机器人刚性物体操作的性能。 |
manipulation cross-embodiment embodiment transfer |
|
|
| 9 |
EfficientFlow: Efficient Equivariant Flow Policy Learning for Embodied AI |
EfficientFlow：面向具身AI的高效等变Flow策略学习 |
manipulation policy learning flow matching |
|
|
| 10 |
$\mathbf{M^3A}$ Policy: Mutable Material Manipulation Augmentation Policy through Photometric Re-rendering |
提出M³A策略，通过光度重渲染增强策略在多材质操作中的泛化能力。 |
manipulation sim-to-real |
|
|
| 11 |
Real-World Robot Control by Deep Active Inference With a Temporally Hierarchical World Model |
提出基于时序分层世界模型的深度主动推理，用于真实机器人控制 |
manipulation world model |
|
|
| 12 |
IGen: Scalable Data Generation for Robot Learning from Open-World Images |
IGen：一种可扩展的机器人学习数据生成框架，利用开放世界图像。 |
manipulation scene understanding |
|
|
| 13 |
Guardian: Detecting Robotic Planning and Execution Errors with Vision-Language Models |
提出自动化机器人故障合成方法以提升故障检测能力 |
manipulation |
|
|
| 14 |
Real-World Reinforcement Learning of Active Perception Behaviors |
提出非对称优势加权回归(AAWR)，解决机器人主动感知行为的现实强化学习问题 |
manipulation reinforcement learning |
✅ |
|
| 15 |
ReGlove: A Soft Pneumatic Glove for Activities of Daily Living Assistance via Wrist-Mounted Vision |
ReGlove：基于腕部视觉的低成本软气动助力手套，辅助日常生活活动 |
manipulation grasping grasp |
|
|