| 1 |
HDCNet: A Hybrid Depth Completion Network for Grasping Transparent and Reflective Objects |
HDCNet:用于抓取透明和反射物体的混合深度补全网络 |
manipulation grasping grasp |
|
|
| 2 |
Unified Humanoid Fall-Safety Policy from a Few Demonstrations |
提出融合人类演示与强化学习的通用人形机器人防摔策略 |
humanoid humanoid robot sim-to-real |
✅ |
|
| 3 |
Dynamics-Decoupled Trajectory Alignment for Sim-to-Real Transfer in Reinforcement Learning for Autonomous Driving |
提出动力学解耦的轨迹对齐方法,实现自动驾驶RL Sim-to-Real零样本迁移 |
sim-to-real motion planning reinforcement learning |
|
|
| 4 |
Time-Aware Policy Learning for Adaptive and Punctual Robot Control |
提出时间感知策略学习,提升机器人控制的自适应性和准时性 |
manipulation sim-to-real reinforcement learning |
|
|
| 5 |
Lightning Grasp: High Performance Procedural Grasp Synthesis with Contact Fields |
提出Lightning Grasp,通过接触场实现高性能程序化抓取合成,加速机械臂操作。 |
manipulation dexterous hand grasp |
|
|
| 6 |
Robot Learning from a Physical World Model |
PhysWorld:通过物理世界建模实现机器人从视频生成中学习 |
manipulation reinforcement learning world model |
✅ |
|
| 7 |
Human-Level Actuation for Humanoids |
提出人形机器人的人体水平驱动评估框架,实现可量化、可比较的性能基准。 |
humanoid humanoid robot walking |
|
|
| 8 |
Exact Smooth Reformulations for Trajectory Optimization Under Signal Temporal Logic Specifications |
提出基于精确平滑重构的轨迹优化方法,解决信号时序逻辑约束下的运动规划问题 |
trajectory optimization motion planning |
|
|
| 9 |
Physically-Grounded Goal Imagination: Physics-Informed Variational Autoencoder for Self-Supervised Reinforcement Learning |
提出物理信息增强的变分自编码器,提升自监督强化学习中目标生成的物理合理性。 |
manipulation reinforcement learning |
|
|
| 10 |
Robotic versus Human Teleoperation for Remote Ultrasound |
对比机器人与人工遥操作超声,评估其在远程医疗中的可行性 |
teleoperation |
|
|
| 11 |
SlotVLA: Towards Modeling of Object-Relation Representations in Robotic Manipulation |
提出SlotVLA框架,用于建模机器人操作中的对象关系表示,并构建LIBERO+基准数据集。 |
manipulation |
|
|
| 12 |
Programmable Telescopic Soft Pneumatic Actuators for Deployable and Shape Morphing Soft Robots |
提出可编程伸缩软气动执行器,用于可展开和变形软体机器人 |
quadruped manipulation |
|
|
| 13 |
Residual Rotation Correction using Tactile Equivariance |
EquiTac:利用触觉等变性进行残差旋转校正,提升操作策略学习的样本效率。 |
manipulation policy learning |
|
|
| 14 |
CAVER: Curious Audiovisual Exploring Robot |
CAVER: 一种好奇心驱动的视听探索机器人,用于构建丰富的物体视听表征。 |
manipulation |
|
|
| 15 |
How Do VLAs Effectively Inherit from VLMs? |
提出GrinningFace基准,诊断VLA模型从VLM有效继承知识的能力 |
manipulation |
|
|