Liwei Hou
Papers
3
Total Citations
14
H-Index
2
About
Liwei Hou is a robotics researcher whose work focuses on the intersection of autonomous manipulation, reinforcement learning, and imitation learning for contact-rich industrial tasks. His primary contributions lie in developing intelligent robotic systems capable of performing high-precision, high-risk operations that were traditionally reserved for human workers. Hou’s most cited work (2022, 8 citations) tackles the challenging problem of automatic peeling of glass substrates for LCD displays, introducing an online learning Model Predictive Path Integral framework to handle delicate contact-rich manipulation with extreme safety requirements. He further advanced robot skill acquisition through a novel policy optimization method (2021, 4 citations) that improves sample efficiency by leveraging weighted near-optimal experiences. Most recently, Hou has pioneered diffusion-based self-supervised imitation learning (2025, 2 citations) to enable robots to learn heavy-duty glass installation tasks from imperfect visual servoing demonstrations, addressing a critical need in modern construction. His research demonstrates a clear trajectory toward making autonomous robots capable of mastering complex, safety-critical manufacturing and construction tasks through data-efficient learning methods.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3