Max Robotics — Global Service Robot Directory + US Certification

About

Ruihan Yang is a robotics and machine learning researcher whose work spans dexterous manipulation, legged locomotion, and humanoid control — areas at the forefront of embodied AI. His research consistently tackles the challenge of enabling robots to learn complex, generalizable behaviors from minimal supervision, often bridging simulation and real-world deployment. Among his most influential contributions is *DexMV* (2022, 117 citations), a framework for learning dexterous hand manipulation by imitating human video demonstrations — a significant step toward scalable robot learning without costly teleoperation. His work on *Expressive Whole-Body Control for Humanoid Robots* (2024, 72 citations) demonstrates how motion capture data can train humanoid policies capable of remarkably lifelike movement. Earlier, his *Soft Modularization* approach (2020, 58 citations) addressed fundamental optimization challenges in multi-task reinforcement learning through dynamic, learned parameter sharing. Yang has also advanced vision-guided quadrupedal locomotion through transformer-based architectures and neural volumetric memory, enabling robots to navigate challenging terrains using depth cameras. More recently, tools like *Bunny-VisionPro* highlight his commitment to practical data collection pipelines for imitation learning. Collectively, his work represents a coherent and ambitious vision: robots that perceive, move, and manipulate with human-level expressiveness and adaptability.