Quadruped Robot Parkour
Abstract Train a quadruped robot parkour policy using RL. The end-to-end policy receives depth image and proprioceptive observation as input and outputs desired joint angles. Demo …
•
1 min read
Selected Projects
I enjoy making things. Here are a selection of projects that I have worked on over the years.
Design and Learning-based Control of Agile-PF： an Agile and Easy-to-Reproduce Point-Foot Biped Robot
Abstract Latest progress in Reinforcement Learning (RL) has spurred growing interest in the development of biped robots. Multiple human-size and small-scale biped robots have …
•
1 min read
Learning Robust Quadrupedal and Bipedal Locomotion on a Quadruped Robot using One Policy
Abstract Locomotion is an essential ability for legged robots to be applied in various conditions. The ability to perform quadrupedal and bipedal locomotion on one robot can bring …
•
1 min read
LeggedGym-Ex
A legged_gym based framework for training legged robots in Genesis, IsaacGym and IsaacSim. This project inherits legged_gym’s framework and brings it to various latest simulators. …
•
1 min read
