Max Robotics — Global Service Robot Directory + US Certification

About

Yashuai Yan is an emerging researcher at the intersection of robotics, deep learning, and human-robot interaction, with a focus on enabling robots to learn and replicate human motion with both physical fidelity and practical applicability. His most recognized contribution, "ImitationNet" (2023, 22 citations), introduced a pioneering unsupervised approach to human-to-robot motion retargeting that eliminates the need for paired training data, significantly lowering the barrier to deploying such systems across diverse robot platforms. Building on this foundation, Yan has advanced the field of humanoid robot control through works like "I-CTRL" (2025), which applies bounded residual reinforcement learning to bridge the gap between visually compelling motion imitation and physics-based execution. His research on push recovery (2024) further demonstrates his commitment to robust real-world bipedal locomotion. Beyond full-body control, Yan has extended his unsupervised imitation framework to dexterous robotic hands, introducing novel cross-domain similarity metrics for fine-grained manipulation. His work on social motion forecasting highlights an additional dimension of his research—making robots socially aware collaborators in human environments. With over 35 cumulative citations across recent publications, Yan is establishing himself as a distinctive voice in embodied AI and humanoid robotics.