Papers
3
Total Citations
32
H-Index
3
About
Wooju Lee is a researcher advancing intelligent robotics through the integration of deep reinforcement learning and multimodal perception. His work centers on developing sample-efficient, occlusion-robust algorithms for robotic manipulation and autonomous navigation. In his highly cited 2022 paper, Lee introduced an adaptive coverage path planning policy for cleaning robots using an actor-critic model and simulator-based training, achieving efficient 2D navigation with deep reinforcement learning (16 citations). He further contributed to outdoor surveillance with the X-MAS dataset, an extremely large-scale multimodal sensor dataset enabling robust human detection and tracking in real environments (10 citations). Most recently, Lee proposed a novel reinforcement learning framework for robotic manipulation that leverages multimodal fusion dualization and representation normalization, significantly improving sample efficiency and robustness to visual occlusions (6 citations). His work bridges simulation and real-world deployment, addressing critical challenges in consumer robotics and autonomous systems. With a growing citation impact, Wooju Lee’s research is shaping the next generation of adaptive, perception-driven robots capable of operating reliably in complex, unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3