Steven Weikai Lu
Papers
2
Total Citations
4
H-Index
2
About
Steven Weikai Lu is a researcher at the intersection of robotics, computer vision, and knowledge representation, with a focus on enabling autonomous systems to learn and reason more effectively. His work centers on two key areas: robot hand-eye coordination learning from demonstration and visual semantic understanding through dynamic knowledge graphs. In his 2019 paper on state representation methods for robot learning, Lu systematically evaluated how different compact representations capture task-relevant information, providing critical insights for improving the efficiency and generalization of imitation learning in robotic manipulation. His second highly cited work introduced a framework for constructing dynamic knowledge graphs that integrate visual perception with common-sense reasoning, enabling autonomous robots to interpret complex scenes and execute tasks with deeper semantic awareness. Though early in his career, with each of his most-cited papers garnering 2 citations, Lu’s contributions lay important groundwork for bridging low-level sensorimotor control with high-level symbolic reasoning—a crucial step toward more intelligent, adaptable robots. His research is particularly relevant for students and researchers exploring embodied AI, human-robot interaction, and knowledge-driven autonomy.
Research Focus
Key Achievements
Top Papers
- 1
- 2