I am currently a doctoral fellow advised by Prof. Marco Hutter and Prof. Andreas Krause at ETH AI Center, Switzerland. Before that, I received my master’s degree in Robotics, Systems and Control from ETH Zurich and my bachelor’s degree from Tongji University.
During my master studies, I worked on structured motion representation and active learning methods at the Biomimetic Robotics Lab at Massachusetts Institute of Technology (MIT) with Prof. Sangbae Kim. Prior to that, I was engaged in agile motion imitation from limited demonstrations and intrinsic skill exploration with Prof. Georg Martius at the Autonomous Learning Group at Max Planck Institute for Intelligent Systems (MPI-IS), Tübingen, Germany.
My research interests focus on algorithms that can enable autonomous agents to acquire complex behaviors through learning, especially general-purpose methods that could enable any autonomous system to learn to solve any task. On legged robots particularly, I am thrilled to apply these techniques to achieve animal-level agility and naturalness.
For more information, please refer to my latest Resumé.
PhD in Robot Learning
ETH AI Center
MSc in Robotics, Systems and Control
BSc in Mechanical Engineering
In this work, we propose a cooperative generative adversarial method for obtaining controllable skill sets from unlabeled datasets containing diverse state transition patterns.
In this work, we propose a generative adversarial method for inferring reward functions from partial and potentially physically incompatible demonstrations for successful skill acquirement where reference or expert demonstrations are not easily accessible.
I Learned and Grew up with
AgnathaX is an undulatory swimming robot equipped with distributed force sensors to study the neural mechanisms controlling locomotion in the spinal cord.
ANYmal is optimized for automated industrial inspection tasks with a wide range of sensors.
ANYmal on Wheels is equipped with non-steerable, torque-controlled wheels to overcome the trade-off between mobility and efficiency.
Mini Cheetah is a lightweight, high-power quadruped robot designed for extraordinarily agile motions.
Solo is an open-source light-weight research quadruped robot that allows highly dynamic actuation.
Spot is an agile mobile robot that navigates complex terrains for automated routine inspection and data capture.