Papers
22
Total Citations
142
H-Index
7
About
Qingdu Li is a robotics researcher whose work spans intelligent robot systems, human-robot interaction, and locomotion control. His research portfolio reflects a broad yet cohesive vision: building robots that move naturally, perceive intelligently, and communicate intuitively with humans. Li has made notable contributions to robot locomotion, from early theoretical work on passive dynamic walking—including a 2012 study on basins of attraction that earned 12 citations—to practical reinforcement learning methods for real-time bipedal gait planning. His 2021 work on table tennis robots addressed the often-overlooked challenge of velocity control, while his innovative spoked Mecanum wheel design demonstrated creative mechanical thinking for multi-terrain mobility. In human-robot interaction, Li has advanced natural language grounding through object affordance detection and scene graph parsing, enabling robots to interpret ambiguous human intent more reliably. His group has also tackled domain adaptation challenges using knowledge distillation techniques, helping robots perceive effectively across changing environments. More recently, Li has explored humanoid expressiveness, publishing work on sign language robots and facial expression imitation. Collectively, his papers have accumulated over 110 citations, reflecting steady and growing recognition across robotics, machine learning, and intelligent systems communities. His research is particularly valuable for students interested in where locomotion, perception, and natural interaction converge.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10