Hong Lu
Papers
7
Total Citations
55
H-Index
4
About
Hong Lu is a robotics and artificial intelligence researcher whose work spans developmental robotics, human-robot interaction, and medical robotics. With foundational contributions beginning in the mid-2000s, Lu helped pioneer biologically inspired learning architectures through the Multilayer In-Place Learning Network, which addressed how developmental robots can incrementally acquire multi-task invariances without predefined programming — work that garnered 24 citations and remains influential in embodied intelligence research. As lead researcher on the FUWA developmental humanoid at Fudan University's Embodied Intelligence Laboratory, Lu demonstrated how autonomous sensorimotor associations could emerge purely from a robot's own experience, advancing the field of developmental robotics. More recently, Lu has bridged robotics and healthcare, developing adaptive sensor fusion techniques and pulse localization networks for robot-assisted Traditional Chinese Medicine diagnosis — innovative work merging computer vision, infrared imaging, and clinical precision. Lu's current research extends into visual place recognition through domain-agnostic contrastive learning and the integration of vision-language-action models with cognitive architectures, addressing critical reliability challenges in generalist robotics. Across more than two decades, Lu's diverse yet cohesive body of work reflects a sustained commitment to building robots that learn, perceive, and interact intelligently with the world.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3DaCo: domain-agnostic contrastive learning for visual place recognition7 citations · 2023
- 4Pulse localization networks with infrared camera7 citations · 2021
- 5The architecture and body of FUWA developmental humanoid4 citations · 2008
- 6
- 7