Papers
2
Total Citations
63
H-Index
2
About
Qunchao Yu is a leading researcher in robotic manipulation, with a primary focus on intelligent grasping systems for unknown and novel objects. His work bridges the gap between computer vision and robotic dexterity, developing deep learning architectures that enable robots to autonomously determine optimal grasp configurations. Yu’s most influential contribution is his novel multilevel convolutional neural network (CNN) framework, which progressively refines grasp detection from coarse orientation to precise finger placement. His 2020 paper on this architecture, extending from parallel grippers to dexterous hands, has garnered 56 citations, marking it as a foundational reference in the field. Earlier, his 2018 work established the core multi-level CNN approach for RGB-D images, achieving efficient and accurate grasp rectangle detection inspired by human grasping strategies. These contributions have significant implications for industrial automation, assistive robotics, and autonomous manipulation in unstructured environments. Yu’s research is distinguished by its systematic, hierarchical approach to a notoriously difficult problem—enabling robots to handle objects they have never seen before—and continues to influence both academic research and practical robotic system design.
Research Focus
Key Achievements
Top Papers
- 1
- 2