Papers
4
Total Citations
113
H-Index
4
About
Yuelin Deng is a leading researcher in robotics and artificial intelligence, whose work bridges the critical gap between theoretical reinforcement learning and practical robotic control. His primary research areas include hierarchical reinforcement learning (HRL), robotic assembly, limbless robot locomotion, and additive manufacturing control. Deng's most impactful contribution is his work on data-efficient HRL for robotic assembly, where he developed algorithms that enable robots to learn complex, multi-step tasks—such as peg-in-hole assembly—with significantly fewer interactions with the environment. This paper has garnered 84 citations, reflecting its influence on making reinforcement learning viable for real-world industrial applications. He has also advanced the field of wire and arc additive manufacturing (WAAM) with a shape-driven control method for layer height, addressing a key challenge in 3D printing on uneven substrates. In a notable departure from assembly-focused work, Deng introduced the "Generalized Omega Turn Gait," a novel turning strategy that dramatically improves the agility of limbless (snake) robots in confined spaces, opening new possibilities for search-and-rescue and inspection tasks. His research is characterized by a strong emphasis on sample efficiency, stability, and generalization, ensuring that his algorithms are not only theoretically sound but also practically deployable.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4