Shengzeng Huo
Papers
14
Total Citations
285
H-Index
7
About
Shengzeng Huo is a robotics researcher whose work spans human-robot interaction, deformable object manipulation, robotic inspection, and safe learning for autonomous systems. His research addresses some of the most challenging frontiers in modern robotics, including enabling machines to handle soft, flexible, and geometrically complex objects with dexterity and intelligence. Huo's most influential contribution — garnering 104 citations — introduces an augmented reality-assisted deep reinforcement learning framework for mutual-cognitive safe human-robot interaction, reflecting his commitment to making collaborative robotics safer and more intuitive. His work on deformable object manipulation is particularly prolific: the LaSeSOM framework (52 citations) proposes a generalizable latent and semantic representation for soft object manipulation, while his keypoint-based bimanual shaping of deformable linear objects (42 citations) offers elegant solutions to high-dimensional manipulation challenges. His research on garment folding using hand-object graph dynamics further demonstrates versatility across manipulation domains. Beyond manipulation, Huo has made notable contributions to robotic surface inspection of specular free-form components, developing sensor-based line-scan systems and PSO-based path planning for quality control in electronics manufacturing. His more recent work on safe reinforcement learning for robotic ultrasound imaging signals a growing interest in medical robotics. Collectively, Huo's research portfolio reflects a rigorous, application-driven vision for intelligent, safe, and capable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10A Robotic Defect Inspection System for Free-form Specular Surfaces5 citations · 2021