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 …

Legged locomotion holds the premise of universal mobility, a critical capability for many real-
world robotic applications. Both model-based and learning-based approaches have …

Parkour is a grand challenge for legged locomotion that requires robots to overcome various
obstacles rapidly in complex environments. Existing methods can generate either diverse …

We present RoboGen, a generative robotic agent that automatically learns diverse robotic
skills at scale via generative simulation. RoboGen leverages the latest advancements in …

Reinforcement learning (RL) for bipedal locomotion has recently demonstrated robust gaits
over moderate terrains using only proprioceptive sensing. However, such blind controllers …

Legged robots are physically capable of navigating a diverse variety of environments and
overcoming a wide range of obstructions. For example, in a search and rescue mission, a …

Recently, humanoid robots have made significant advances in their ability to perform
challenging tasks due to the deployment of Reinforcement Learning (RL), however, the …

Deep reinforcement learning can enable robots to autonomously acquire complex behaviors
such as legged locomotion. However, RL in the real world is complicated by constraints on …

We introduce Berkeley Humanoid, a reliable and low-cost mid-scale humanoid research
platform for learning-based control. Our lightweight, in-house-built robot is designed …

We investigate the use of transformer sequence models as dynamics models (TDMs) for
control. We find that TDMs exhibit strong generalization capabilities to unseen …