Hanzhi Chen

Papers

Total Citations

H-Index

About

Hanzhi Chen is an emerging robotics researcher whose work sits at the intersection of robot manipulation, computer vision, and autonomous exploration. His research focuses on enabling robots to operate more intelligently in human-centered environments, with particular emphasis on dexterous grasping, learning from human demonstration, and autonomous navigation. Chen's most recognized contribution, "Anthropomorphic Grasping With Neural Object Shape Completion" (2023, 15 citations), addresses a fundamental challenge in robotics: replicating the extraordinary dexterity humans naturally exhibit when handling objects. By leveraging neural shape completion, his approach allows robots to grasp objects more human-like, even under partial observability — a significant step forward for real-world manipulation tasks. His more recent work broadens this vision considerably. "VidBot" (2025) proposes an innovative framework for extracting generalizable 3D actions from everyday 2D human videos, enabling zero-shot robotic manipulation without expensive physical robot training. Meanwhile, "FrontierNet" (2025) advances autonomous exploration by incorporating learned visual cues into frontier-based navigation strategies. Together, these contributions reflect a cohesive research agenda: reducing the gap between human capability and robotic performance through scalable, vision-driven learning — positioning Chen as a promising voice in next-generation intelligent robotics.