Chao Yu is a research-driven engineering manager with eight years’ experience leading legged-robot and humanoid locomotion teams, currently heading the leg robot group at Dreame Technology in Shanghai. He blends deep reinforcement learning, multi-agent RL and vision-based grasping research with practical product delivery, having driven walking, manipulation and perception algorithms for biped and legged platforms. A Tsinghua-trained engineer (MEng, aerospace; BS in Mathematics and BE in Energy/Automation), he has hands-on contributions to notable open-source MARL tooling—improving MAPPO implementations for StarCraft II and multi-agent particle environments with value normalization and rendering support. Colleagues describe him as someone who turns cutting-edge embodied-AI ideas into robust, testable systems and who keeps one foot in academia while shipping industrial robots.
8 years of coding experience
Master of Engineering - MEng, Aeronautics/Aviation/Aerospace Science and Technology, General, Master of Engineering - MEng, Aeronautics/Aviation/Aerospace Science and Technology, General at Tsinghua University
This is the official implementation of Multi-Agent PPO (MAPPO).
Role in this project:
ML Engineer
Contributions:1 review, 55 commits, 19 PRs in 1 year 8 months
Contributions summary:Chao contributed to the Multi-Agent PPO (MAPPO) implementation by modifying code related to environment interactions, specifically updating environment definitions for StarCraft II and MPE environments. They added code to support value normalization and PopArt, as well as rendering functionality for MPE environments. The user also made bug fixes and configuration updates related to training and rendering.
Contributions:132 commits, 139 pushes, 1 branch in 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.