Yuxuan Kuang
Papers
2
Total Citations
4
H-Index
2
About
Yuxuan Kuang is a rising roboticist whose work bridges perception and manipulation, tackling some of the field’s most stubborn real-world challenges. His research centers on two critical areas: robust object detection for industrial automation and generalizable, data-efficient robotic manipulation. In his 2024 paper “STOPNet,” Kuang addresses the notoriously difficult problem of 6-DoF suction detection for transparent objects on production lines—a task where standard depth sensors fail. By leveraging multiview inputs, his framework achieves reliable grasping of these visually challenging items, directly impacting modern manufacturing efficiency. Complementing this, his work “RAM” introduces a retrieval-based affordance transfer framework for zero-shot robotic manipulation. This innovative approach allows robots to generalize manipulation skills across novel objects, environments, and even different robot embodiments without requiring expensive in-domain demonstrations. Though early in his career, Kuang’s contributions are already garnering attention, with each paper accumulating citations that signal growing influence. His focus on practical, scalable solutions—from transparent object handling to cross-embodiment transfer—positions him as a key voice in the next generation of robotics research, where perception and action must work seamlessly together.
Research Focus
Key Achievements
Top Papers
- 1
- 2