Zhengmao He

Papers

Total Citations

H-Index

About

Zhengmao He is an emerging robotics and artificial intelligence researcher whose work sits at the intersection of reinforcement learning, robot manipulation, and embodied intelligence. His research tackles some of the most challenging problems in modern robotics: enabling machines to perceive, reason, and act effectively in complex, real-world environments. He is perhaps best known for ArrayBot, his innovative distributed manipulation system featuring a 16×16 array of tactile-sensing pillars capable of simultaneously supporting, perceiving, and manipulating tabletop objects. This work, which has garnered 12 citations since its 2024 publication, represents a significant advance in generalizable robotic manipulation through touch-based reinforcement learning. His parallel work on visual quadrupedal loco-manipulation — teaching legged robots to interact with objects through demonstration-based learning — has attracted 11 citations, underscoring his broad influence across locomotion and manipulation domains. Beyond hardware-grounded robotics, He has also contributed to foundational reinforcement learning methodology, proposing Uni-O4, a unified framework bridging offline and online deep RL through multi-step on-policy optimization. Though an early-career researcher, He's diverse and rapidly cited body of work signals a promising trajectory in next-generation autonomous robotic systems.