Kelin Yu
Papers
2
Total Citations
57
H-Index
2
About
Kelin Yu is a leading researcher at the intersection of robotics, tactile sensing, and dexterous manipulation, with a focus on enabling machines to handle complex, contact-rich tasks. Their most-cited work, "Learning Generalizable Vision-Tactile Robotic Grasping Strategy for Deformable Objects via Transformer" (2024, 54 citations), addresses the fundamental challenge of reliably grasping deformable objects like fruits. By proposing a transformer-based framework for rigid grippers, Yu tackles the complexities of underactuated contact and unknown object dynamics, offering a generalizable solution that bridges vision and touch. In their more recent study, "MimicTouch: Leveraging Multi-modal Human Tactile Demonstrations for Contact-rich Manipulation" (2023, 3 citations), Yu pioneers a novel approach to learning tactile-guided policies. Recognizing that human demonstrators often lack direct tactile feedback, MimicTouch leverages multi-modal human demonstrations to teach robots fine-grained skills like insertion and assembly. This work is notable for its potential to revolutionize how robots learn from humans in manufacturing and healthcare. With a growing citation impact and a focus on real-world applicability, Kelin Yu is shaping the future of robotic manipulation, making it safer, more intuitive, and more capable in unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2