A general RL framework for bipedal robot to perform various locomotion skills from standing to walking, jumping, and running.

A bipedal robot now can autonomously and safely navigate unknown and congested environments.

Learn from real-world RGBD visual data to navigate with lower traversability cost for quadrupedal robots.

A legged robot team that can tow a heavy load by cables.

A decentralized collaborative quadupedal robots team for load-towing tasks by MARL.

An adpative control framework for interactive navigation tasks.

Soccer team play in real!

A fully open-source 3D-printed humanoid robot!

Text-To-Humanoid-Control

LLM agents for long-horizon task and motion planning for loco-manipulation tasks using quadrupeds

Using Lipschitz constraint to have smooth RL policy.

An in-house-built and cool humanoid (bipedal) robot.

LLM agents for curricula design for RL.

Diffusing multimodal legged locomotion skills purely from offline datasets with a 6.8M transformer DDPM (YES it runs onboard at 50Hz!).

Imitation learning for end-effector planning + reinforcement learning whole-body contol for loco-manipulation tasks.

The efficacy of two approaches to incorporate symmetry in RL for locomotion control.

A hierarchical RL framework to tackle various loco-manipulation tasks while performing continuous quadrupedal locomotion skills.

A bipedal robot learns to perform various aggressive jumping maneuvers.

Iterative LQR for iterative tasks, analogous to learning MPC :)

A safety-critical legged locomotion controller to travel through narrow spaces.

A agile quadrupedal goalkeeper that can intercept fast-moving ball.

An opensourced generalized locomotion controller for a vast number of quadrupedal robots.

A quadrupedal robot that can precisely shoot a soft ball to the given target.

An adaptive bipedal locomotion controller by RL.

Humans can teach robot how to behave with emotions!

A bipedal robot controlled by its RL policy is a linear system!

Auto-tuning a HZD-based bipedal locomotion controller by Bayesian Optimization.

Quadrupedal robots can traverse more complex environments by leveraging a big jump.

Incorporating Control Barrier Functions (CBFs) with Model Predictive Control (MPC).

A bipedal robot learns to walk, turn, and squat, with significant robustness.

Quadrupedal robots can serve as guide dogs!

How can we ensure feasibility using Control Barrier Functions (CBFs) with bounded inputs?

A bipedal robot can behave emotively to be a better human companion!

A ballbot that can lead a blind-folded person to navigate environments. This is how my story begins.