Wenli Xiao

I am a master student at CMU Robotics Intitute (MSR) advised by Prof. Guanya Shi and Prof. John Dolan.

I am currently doing research intern at NVIDIA GEAR Lab working with Dr. Jim Fan and Prof. Yuke Zhu on foundation model for humanoids.

Email  /  Google Scholar  /  Twitter  /  Github

TL;DR: HOVER is a 1.5M-parameter neural network to control the body of a humanoid robot. It takes a lot of subconscious processing for us humans to walk, maintain balance, and maneuver our arms and legs into desired positions. We capture this “subconsciousness” in HOVER, a single model that learns how to coordinate the motors of a humanoid robot to support locomotion and manipulation.

TL;DR: AnyCar is a generalist vehicle dynamics model for agile mobility. It can adapt to various cars, tasks, and envs via in-context adaptation, outperforming well-tuned generalist models up to 54%.

TL;DR: WoCoCo is the first unified RL framework to learn whole-body humanoid control with sequential contacts.

TL;DR: OmniH2O provides the first universal whole-body humanoid control interface that enables diverse teleoperation and autonomy methods.

TL;DR: H2O enables real-time whole-body teleoperation of a full-sized humanoid to perform tasks like pick and place, walking, kicking, boxing, etc.

This paper jointly tackles policy adaptation and safe reinforcement learning with safety guarantees. Comprehensive experiments on (1) classic control problems (Inverted Pendulum), (2) simulation benchmarks (Safety Gym), and (3) a real-world agile robotics platform (RC Car) demonstrate great superiority of SafeDPA in both safety and task performance, over state-of-the-art baselines.

Legged robots navigating cluttered environments must be jointly agile for efficient task execution and safe to avoid collisions with obstacles or humans. Existing studies either develop conservative controllers (< 1.0 m/s) to ensure safety, or focus on agility without considering potentially fatal collisions. This paper introduces Agile But Safe (ABS), a learning-based control framework that enables agile and collision-free locomotion for quadrupedal robots.