Reinforcement learning (RL), particularly its combination with deep neural networks,
referred to as deep RL (DRL), has shown tremendous promise across a wide range of …

Embodied Artificial Intelligence (Embodied AI) is crucial for achieving Artificial General
Intelligence (AGI) and serves as a foundation for various applications that bridge cyberspace …

We introduce UMI-on-Legs, a new framework that combines real-world and simulation data
for quadruped manipulation systems. We scale task-centric data collection in the real world …

Humanoid robots capable of autonomous operation in diverse environments have long
been a goal for roboticists. However, autonomous manipulation by humanoid robots has …

Recent works have been exploring the scaling laws in the field of Embodied AI. Given the
prohibitive costs of collecting real-world data, we believe the Simulation-to-Real (Sim2Real) …

The ability for robots to perform efficient and zero-shot grasping of object parts is crucial for
practical applications and is becoming prevalent with recent advances in Vision-Language …

Reinforcement learning combined with sim-to-real transfer offers a general framework for
developing locomotion controllers for legged robots. To facilitate successful deployment in …

Learning-based methods have achieved strong performance for quadrupedal locomotion.
However, several challenges prevent quadrupeds from learning helpful indoor skills that …

Flying through body-size narrow gaps in the environment is one of the most challenging
moments for an underactuated multirotor. We explore a purely data-driven method to master …

Recent advances in diffusion models have opened new avenues for research into embodied
AI agents and robotics. Despite significant achievements in complex robotic locomotion and …