| 1 |
Seeing the Bigger Picture: 3D Latent Mapping for Mobile Manipulation Policy Learning |
提出基于3D隐空间地图的移动操作策略学习方法，增强空间和时间推理能力。 |
manipulation mobile manipulation reinforcement learning |
|
|
| 2 |
Learning to Act Through Contact: A Unified View of Multi-Task Robot Learning |
提出基于接触的统一多任务机器人学习框架，实现通用运动与操作策略 |
quadruped humanoid biped |
|
|
| 3 |
NoTVLA: Narrowing of Dense Action Trajectories for Generalizable Robot Manipulation |
提出NoTVLA框架，通过稀疏轨迹学习解决VLA模型中的灾难性遗忘问题，提升机器人操作的泛化性。 |
manipulation vision-language-action VLA |
|
|
| 4 |
An Amphibious Untethered Inchworm Soft Robot for Fast Crawling Locomotion |
提出一种新型无束缚磁驱动尺蠖软体机器人，实现快速爬行运动 |
locomotion multimodal |
|
|
| 5 |
EmbodiSwap for Zero-Shot Robot Imitation Learning |
EmbodiSwap：利用合成数据实现机器人零样本模仿学习 |
manipulation imitation learning Ego4D |
|
|
| 6 |
COVER:COverage-VErified Roadmaps for Fixed-time Motion Planning in Continuous Semi-Static Environments |
COVER：面向半静态环境固定时间运动规划的覆盖验证路径图 |
motion planning |
|
|
| 7 |
Model-Based Adaptive Precision Control for Tabletop Planar Pushing Under Uncertain Dynamics |
提出基于模型的自适应精度控制方法，解决不确定动力学下的桌面平面推移任务。 |
sim-to-real domain randomization |
|
|